Product Hunt 每日热榜 2026-01-15

PH热榜 | 2026-01-15

#1
Cal.com Companion Apps
Native iOS & Android App + Browser Extensions for scheduling
476
一句话介绍:Cal.com推出原生iOS与Android应用及浏览器扩展,在移动和网页场景下,为用户提供免费、跨平台、体验统一的日程安排解决方案,解决了用户在外出时快速分享预约链接和管理日程的痛点。
Android Productivity Calendar Apple
日程安排工具 原生移动应用 浏览器扩展 跨平台 开源 免费替代品 生产力工具 用户体验 代码复用 团队协作
用户评论摘要:用户普遍对原生应用发布感到兴奋,认为这是产品缺失的关键一环。主要反馈包括:期待深色模式、询问Apple Watch支持、赞赏其作为Calendly的免费开源替代品。技术社区对其使用单一代码库(Expo+WXT)实现多平台部署的工程能力表示高度认可。
AI 锐评

Cal.com此次发布远非简单的功能堆砌,而是一次精准的战略卡位。在Calendly等巨头垄断的日程管理市场,它打出了一套差异化组合拳:以“全平台免费原生体验”直击付费产品的软肋,用“开源+API优先”吸引开发者和企业级用户,再以单一代码库的高效工程实践展示技术肌肉,构建起从个人到企业的完整叙事。

然而,光鲜背后暗藏挑战。其一,“全平台原生”的口号与使用React Native/Expo等技术实现的“跨平台原生”之间存在体验鸿沟,评论中关于“最难妥协的UX设计”的提问恰恰点中了这一潜在矛盾。真正的原生体验绝非一套UI组件库所能涵盖。其二,在巨头环伺的赛道,免费开源模式如何持续造血?这不仅是商业问题,更关乎其能否长期维持如此高强度的多平台同步开发与迭代。用户已开始索取深色模式、手表支持等“痒点”功能,这对其资源分配是持续考验。

其真正的价值或许不在于“又一个日程安排工具”,而在于它试图成为调度领域的“基础设施”。通过提供从自托管到云服务、从API到终端用户应用的全栈解决方案,它正从工具向平台演进。此次移动端补齐,是其完善生态闭环的关键一步,旨在将用户彻底锁定在自己的体系内。成败关键在于,它能否在快速扩张中,守住“体验统一”与“开发效率”的脆弱平衡,避免沦为又一个功能齐全但体验平庸的“缝合怪”。

查看原始信息
Cal.com Companion Apps
After 4 years and many customer requests we are finally launching our native iOS and Android apps + Browser Extensions for Google Chrome, Firefox, Safari and Edge

WOW what a ride. It's finally time for us to launch our native iOS and Android apps AND chrome, firefox and safari browser extensions.

this has been in the making for quiet some time and we're super excited to be the only **free** scheduling app with native apps in every store.

our goal has always been to deliver the best experience on any platform we launch and we finally found time and resources to invest in our mobile experience.

the first v1.0 launches today but you can be excited for widgets, live activities, push notifications coming up very soon!

let us know in the comments how you like the apps and what else is missing 🙏

the iOS Apps are using the latest liquid UI which looks and feels really neat, on Android we deliver a native Material Design look and feel and the browser extensions really fit in nicely as a sidekick in your daily browser usage.

we've added a ton of cool tricks in the extension, for example you can now inject your links into mail.google.com or mark people as "no-show" inside calendar.google.com

fun fact: Both mobile apps and browser extensions are built with the same codebase (expo + WXT, read more how we built it here: https://cal.com/blog/how-cal.com...)

i hope you are as equally excited as i am about this launch

peer & the team

19
回复

@peer_rich congrats on the launch of the mobile apps. The coolest scheduling app by far!

0
回复

@peer_rich Shipping native apps and extensions from one codebase is impressive. What was the hardest UX compromise you had to make to keep the experience feeling truly native on iOS, Android, and the browser at the same time?

0
回复

@peer_rich love cal.com and love this, been using for a couple of years now and this is so clutch!

1
回复

I discovered you guys a couple of years ago and Cal was one of the reasons for me to go build in OSS (let alone to forget about Calendly). Glad now it takes me less time to get my link while on the go

But – where's the dark theme? hahahahahah

4
回复

@vlad_nadymov LETS GOO glad you like it. yeah no darkmode yet. hopefully soon

1
回复
After recently setting up Cal for my startup and being surprised by how seamless the experience was, this mobile app was the missing piece. Congrats to the whole team!
4
回复

@tonyystef glad you like it! let us know if you think anything is missing

1
回复

FINALLY - I had to do a double take to see if this was an official launch. Thank you @peer_rich & cal team for finally brining this to mobile TT__TT

4
回复

@gabe yess!! finallly. way overdue

2
回复

Congrats on the launch! Love how Cal.com brings open, API-first scheduling infra to everyone—self-hosted or hosted—without sacrificing customization, privacy, or enterprise readiness.​

3
回复

Huge!!! Congrats team! 🔥

2
回复

@didier_lopes thank you didier

0
回复

Fantastic! Really appreciate the effort in making it truly native to iOS AND Android (often left out). Thank you!

1
回复

Lmao, I love the thumbnail on the video

1
回复

@skeptrune ahha thanks!

0
回复

It’s rare to see a product launch that actually makes daily life simpler. This one does. Huge congrats! Quick question: any plans for Apple Watch support?

1
回复

@abod_rehman hmm how is apple watch support on expo? https://github.com/EvanBacon/expo-apple-targets

Lemme check. could be an option. also a reason for me to finally get an Apple Watch haha

0
回复
Let’s goooo! Congrats on the launch!
1
回复

@davj YEEEEE

0
回复

have been looking for a alternative to calendarly for long... hope Cal could help me out!

1
回复

@cruise_chen lets do it!

0
回复
When someone already has a scheduling link tool (or uses Google/Microsoft’s built-in booking features), what are the specific switching triggers you expect and which of those is proving most compelling in early adoption?
1
回复

The Cal.com engineering team is so cracked! A single React Native Expo app shipped to every platform, and they built it in just weeks! It’s crazy what small teams of highly skilled engineers can do. Congrats on the launch!

1
回复

@betomoedano ayyyyy

0
回复

Congrats Peer and team! Can't wait to use the app 🤙

1
回复

@airbnbforsurfboards thank you joe 🫡

0
回复

An amazing feat of engineering, can't wait to use it, will make my scheduling workflow much better.

1
回复

@daniel_barankin thanks daniel

0
回复

Wow, Cal.com looks incredible! Love the open source approach and native apps. Im curious, how well does the open API handle complex conditional logic for meeting types? Congrats on the launch!

0
回复

seriously if you're using calendly, just switch to cal it is literally better in every way and they're actually building and improving the product

0
回复

This is HUGE, have been waiting the Cal.com iOS app since forever!

0
回复

@steventey YEEEEE -- wait until you know what's behind go.cal.com/ph 👀

0
回复
#2
TeamOut AI
The AI that plans your events
399
一句话介绍:TeamOut AI是一款企业活动策划AI副驾,通过对话式界面为远程团队和初创公司智能推荐场地、估算预算并获取专属折扣,将繁琐的团队差旅和线下活动策划从数天压缩至几分钟。
Productivity Travel Artificial Intelligence
AI企业服务 活动策划 SaaS 对话式AI 差旅管理 团队建设 企业采购 智能推荐 流程自动化 成本优化
用户评论摘要:用户肯定其用AI推理替代传统筛选的颠覆性体验,认为解决了活动策划中预算、航班、场地等多约束协调的核心痛点。主要建议包括:增加地理位置识别、提供灵感推荐以防用户无明确方向、关注AI在对话中实时适应需求变化的能力。
AI 锐评

TeamOut AI的亮相,与其说是一款新工具,不如说是一次对“AI-ready”场景的精准狙击。它避开了已被做烂的C端旅行规划,直击企业活动策划这个“预算敏感、多约束、强沟通”的修罗场。其宣称的“对话式规划”并非简单的聊天机器人,而是试图将项目经理、行政、差旅采购的多角色对话与决策流程封装进一个多智能体系统。这才是其真正的价值主张:不是信息聚合,而是流程再造。

然而,其面临的挑战同样尖锐。首先,企业采购流程复杂,AI生成的“24小时内报价”能否无缝对接企业内部的审批、法务、报销链条?这涉及系统之外的生态整合。其次,其核心优势“专属折扣”建立在房源合作深度上,这更像是一个商务能力,而非纯粹的技术壁垒,容易被后来者复制。最后,也是最关键的,是将高度非标、充满人情世故的“活动策划”抽象为可被AI理解的结构化约束,其边界在哪里?用户评论中关于“中途优先级变更”的疑问,恰恰点中了当前AI在处理复杂、动态人类协作时的软肋。

产品团队背景(前IBM AI研究员)赋予了技术可信度,但最终胜负手在于:能否将“理解意图”转化为“闭环交易”。如果它仅停留在智能推荐层面,则只是一个更聪明的过滤器;若能真正打通从聊到订、再到后续变更的全流程,才算真正解锁了新 workflow。当前版本是一个极具潜力的楔子,但楔入企业厚重流程的深度,将决定其是成为必需品,还是又一个效率玩具。

查看原始信息
TeamOut AI
TeamOut is the first AI copilot for planning team offsites and corporate events. Instantly get tailored property recommendations, flight cost estimates, and exclusive hotel deals that save you up to 40 percent. No waiting, no planners, just smarter planning in minutes. Whether you are organizing a company retreat or a team gathering, TeamOut helps you go from idea to booked faster and easier than ever before. Perfect for remote teams and fast-moving startups.

👋 Hi Product Hunt, and thank you @garrytanfor hunting us 🚀


I’m Vincent, Co-founder & CTO of TeamOut AI.

TeamOut was founded in 2021 because of a problem we experienced ourselves: corporate event planning is one of the most painful tasks in companies. Budgets, flights, venues, AV, availability, trade-offs… and endless back-and-forth. It’s repetitive, highly conversational, slow, and incredibly fastidious.


In other words: it was always meant for AI.

Before TeamOut, I worked as an AI researcher at IBM, focusing on natural language processing and recommendation systems, and authored scientific papers in applied AI.

For years, we knew this category wanted an AI copilot but the technology simply wasn’t ready. Filters and forms can’t reason. Early AI couldn’t hold real planning conversations.

Now it can!

We’re finally at the intersection of market ready + technology ready.

And that unlocked what the event planning world had been waiting for.

We built the Lovable of Event Planning, because that’s the revolution this category needed. Not another marketplace. Not better filters. A product you talk to, that understands constraints, asks the right questions, and adapts as plans evolve.


TeamOut AI is the first AI copilot for corporate event planning.

You plan by talking:

  • Compare destinations

  • Estimate full event & flight budgets

  • Check meeting space, AV, and constraints

  • Get instant recommendations from vetted venues where real teams have stayed

Under the hood, it’s a fully custom interface powered by a multi-agent AI system, combining conversational reasoning, recommendation models, and real operational data.

What makes TeamOut AI different:

  • Conversational planning instead of filters

  • Instant, contextual recommendations

  • Quotes in under 24 hours

  • Built-in budget & flight estimators

  • Preferred rates up to 30% below public pricing

The technology is finally ready, and we finally built what the event planning world has been asking for.


Would love your feedback. Try it, push it, ask it hard questions. 🚀

Thanks for checking it out 🙏

18
回复

@garrytan  @vincalbouy Congrats on the launch! Can the copilot adapt mid-conversation when priorities change? Interesting to test this one. 👀

3
回复

@garrytan  @vincalbouy Event planning is all constraints and conversations, not filters.

Talking your way through destinations, budgets, and trade-offs feels like the right UX for this category. The timing argument also lands. This was always an AI problem, just not solvable until now.

I see the same shift while building Curatora. When AI can reason and adapt instead of just retrieve options, whole workflows unlock. Excited to see teams actually stress-test this with real planning chaos.

6
回复

@garrytan  @vincalbouy Guys ! Great idea :) eEvent planning always breaks when reality hits. What’s the first constraint that usually forces teams to rethink everything mid-conversation, budget drift, flight availability, internal approvals, or last-minute headcount changes ?

1
回复

One of my responsibilities is planning and booking flights and all kinds of tickets to conferences for members of our company and your product is right on point for these purposes. Dealing with flights, WiFi speed, and all this typa stuff is a logistical nightmare. AI that actually reasons through these constraints instead of just showing me a list of hotels is huge. So ty and congrats on the launch! Does the budget estimator include local transport, or just flights and venue?

5
回复

@eugene_chernyak Thanks so much — that’s exactly the pain we built TeamOut AI for 🙏
You’re spot on: flights + WiFi + availability + timing turns into a logistical nightmare fast, and lists alone don’t help.

On the budget estimator: today it covers flights and the venue (including meeting space & AV). Local transport is also included YES.

Really appreciate the kind words — and congrats to you for surviving conference logistics 😄🚀

2
回复

Hey PH, I am Thomas, the other co-founder of TeamOut. Today is a bit of an emotional moment for us because it marks the culmination of four years of hard work.

TeamOut started as just an Airtable list of top hotels in California. That was our v0.1 back in 2021. From there, we kept iterating by building and gathering feedback from companies like Reddit, OpenAI, Netflix, and more. Today, seeing what this relatively small team has been able to put together, and the level of quality we reached, honestly made me tear up a little.

It is truly incredible what a small team can achieve when they find the right tools and push hard in the same direction. What a day.

Thank you to everyone who gave our product a chance and helped us make it better every day.

2
回复

@thomas_mazimann2 This one really hits. Seeing how far we’ve come from those early days is wild. Immensely proud of the team and grateful to everyone who trusted us along the way. From an Airtable list to what we launched today the journey has been unreal

0
回复

Hey everyone 👋

I'm Julien, part of the TeamOut team and one of the engineers behind TeamOut AI.

We built this new product to make it much easier for teams to discover the right venues for their events. This launch is just a first step, and we're excited to learn from your feedback.

Happy to answer any questions 🙌

2
回复

@lildesert Super proud of what we achieved here, Julien. This launch is the result of an impressive level of commitment and dedication over many months, and it really shows in the final product.

What stands out most is how well you navigated a constantly evolving AI landscape while staying grounded in real product constraints. On top of that, your strong sense of UI and UX ensured the experience stayed clear, intuitive, and genuinely useful, not just technically impressive.

The speed of execution, the quality bar you maintained, and your ability to adapt and iterate quickly made a huge difference. Thanks for pushing this forward with so much ownership and care. Excited to see where you take this next 🙌🚀

0
回复
Congrats on the launch 👍 Nice idea and, indeed, it is innovative with the search/chat interface where users can provide input in plain English. As a feedback, since the app is so powerful and I almost had a similar idea, I believe that the app should use geolocation. Another aspect is: what if the user doesn’t really know where to head or what event theme they want. I believe the app could benefit from inspirational listings or at least a choice provided in the chat interface. This is just my perspective, but of course you know your product better
2
回复

@zahran_dabbagh Thank you so much for this it really means a lot !
This journey has been a constant balance between moving fast, adapting to rapidly evolving AI tech, and staying focused on building something genuinely useful for teams. Hearing that the UI, UX, and overall product direction resonate makes all the effort worthwhile.

I’m grateful for the trust, the support, and the opportunity to build this together. This launch is just the beginning, and I’m excited to keep iterating and pushing the product forward .

1
回复

Love this,planning offsites finally feels effortless. So… do I still need to pack my own suitcase or is that coming in v2? 😄

2
回复

@mees_schiks V2 will definitely help you to choose the right color of swimsuit for your next retreat in Tulum 🏖️

2
回复

Congrats on the launch! Love how TeamOut makes “Airbnb for retreats” actually real—large, WiFi-ready spaces plus streamlined planning so teams can focus on connection instead of logistics.​

2
回复

@zeiki_yu  Thanks so much, Zeiki — really appreciate that 🙏
That’s exactly what we’re aiming for: making team retreats feel effortless, so teams can focus on connection instead of logistics.
The “Airbnb for retreats” idea was the starting point, and the AI copilot is what finally makes it work end-to-end. Thanks for the support! 🚀

1
回复

I really love products built by people who were personally traumatized by a problem :D. This one definitely shows that energy.

1
回复

congrats on the launch!!

1
回复

Big fan of TeamOut! Congrats for the launch!

1
回复

@qhoang09  Thank you Quang for the infinite support !

0
回复

As longtime event planner, now on the TeamOut team, this is the tool I wish I had years ago!

So much of corporate planning is spent chasing specs, sanity-checking budgets, and answering the same questions over and over. TeamOut AI finally removes that friction so planners can focus on strategy, creativity, and experience. Proud of what the team built 🙌

1
回复

Hi all 👋


I’m Lorea from TeamOut, part of the team that works directly with companies planning offsites.

TeamOut AI came out of a lot of real planning pain we kept seeing over and over. We’re excited to see how people use it and would love any thoughts or questions.

1
回复

finally can find a place for our team building! Nice idea Teamout!

1
回复

@cruise_chen Love to hear that! 🎉 Finding the right team-building spot is usually the hardest part — excited for you to try TeamOut. Happy to help if you need anything.

0
回复

We are excited to try this feature. We used TeamUp before for our team retreats on Prefix, and we would love to see how it works.

1
回复

@copaceticd Thanks Dan, really appreciate that! 🙌 Excited for you to try it and hear how it compares to your past TeamUp retreats at Prefix. Would love your feedback as you explore TeamOut AI.

0
回复

Hey everyone 👋

I’m Sofia, part of the TeamOut team, working on Venue Partnerships & Business Development.

Super proud to be part of this launch. If you’ve ever planned a corporate event, you know how broken and time-consuming the process can be. We truly believe AI is the missing piece here, and we can’t wait to hear your feedback 🚀

1
回复

congrats the team for the release!

1
回复

@simoelalj  Thank you for the support. It all started from a blue house in San Francisco

1
回复

Hype to be part of the team that’s out here making it all make sense! Pricing, but make it instant. Actually feels like a convo. No digging through endless venue tabs. Tbh, this is how planning should feel.

1
回复

This is genius for teams. I dig the idea.

1
回复

@deduxer Thanks so much Roma! Love what you're building at Accordio AI btw, happy to chat about team retreats anytime!

1
回复

Congrats on the launch, looks awesome !

1
回复

@nicolas_laporte  Thank you so much for your help and support

0
回复

We used it to organise a team retreat (12 people) in Colombia this was amazing!!

0
回复

Wow, TeamOut looks amazing! I love the AI copilot aspect - sounds like a massive time saver. How does the algorithm handle different team sizes to ensure the recommendations are always relevant?

0
回复

What happens when a recommendation looks good but turns out to be a poor cultural or experiential fit for the team? Is there a way we can make AI learn from that feedback??

0
回复

Congrats on the launch!! Does it capture real-time accurate data for flight costs and hotel deals?

0
回复

I’ve just taken a brief look and found myself genuinely curious: what would you say is the key differentiation between this product and platforms like Airbnb or booking? Is the core innovation mainly centered around using AI for user understanding, information extraction, and qualification?

0
回复

It made me think how will it handle distributed or global teams like ours. How does the AI handle visa constraints, time zones and travel complexity without oversimplifying the plan?

0
回复

Once a plan is generated, can teams flexibly iterate, negotiate or customize deeply before actually booking everything?

0
回复

What makes TeamOut meaningfully different from working with a human event planner or using tools like Google Sheets, travel sites, and Slack together?

0
回复

Hey everyone 👋
I’m Sofia, part of the TeamOut team, working on Venue Partnerships & Business Development.
Super proud to be part of this launch. If you’ve ever planned a corporate event, you know how broken and time-consuming the process can be. We truly believe AI is the missing piece here, and we can’t wait to hear your feedback 🚀

0
回复

team off-sites are one of those things everyone knows are valuable, yet planning them is always way more painful than it should be. For remote and fast-moving teams, this is exactly the kind of tool that turns a “we should do an offsite” idea into something that actually gets booked. Congrats on the launch 🙌

0
回复

There are soooo many tasks that you always feel are completely non value added when planning an offsite yet super time consuming and mandatory (coordinating flights, food etc..)
Glad someone is giving a try at fixing this with AI!
Nice work les gars @vincalbouy @thomas_mazimann2

0
回复
#3
folk Assistants
Sales Assistants working 24/7 to help you close more deals
277
一句话介绍:一款将AI深度嵌入CRM系统的智能助手,通过自动研究客户、发送外联邮件、生成交易复盘和识别待跟进对象,在销售全流程中主动执行任务,解决了传统CRM依赖人工更新、操作繁琐且被动的核心痛点。
Sales Artificial Intelligence CRM
AI销售助手 智能CRM 销售自动化 客户关系管理 工作流自动化 主动式CRM 销售效率工具 B2B销售 AI赋能 SaaS
用户评论摘要:用户反馈积极,认可产品将AI深度嵌入工作流的理念。主要问题集中于:与Linkedin等外部工具集成能力、AI判断行动优先级的逻辑、以及相比其他CRM“副驾驶”功能的端到端执行优势。建议包括对“即将推出”功能给予更多提示。
AI 锐评

folk Assistants的发布,并非简单地为CRM增加一个AI聊天机器人,而是试图对CRM的底层逻辑进行一次“范式革命”。其真正的价值在于将CRM从“被动记录系统”重新定义为“主动执行系统”。传统CRM的本质是一个需要人工喂养和操作的数据库,其价值与数据录入的及时性和完整性强相关,这恰恰导致了极高的使用疲劳和数据滞后。folk提出的“像拥有一个助理团队”的愿景,直指这一核心矛盾。

从评论中团队对“端到端执行”的强调可以看出,其产品野心是让AI成为工作流的驱动引擎,而非仅仅是查询界面。研究、外联、复盘、跟进这四大助理,覆盖了从线索挖掘到成交后维护的销售生命周期关键节点,旨在构建一个闭环的自动化销售机器。这比当前市场上主流的“AI总结摘要”或“智能问答”式CRM辅助功能更为激进和彻底。

然而,其面临的挑战同样清晰。首先,是AI行动的“判断力”与“可控性”之平衡。系统如何精准识别“值得行动的信号”而非制造干扰,是决定用户体验成败的关键。其次,深度自动化在提升效率的同时,也可能带来个性化缺失和沟通温度下降的风险,这与CRM“专注于关系”的初衷存在微妙张力。最后,在数据安全与隐私监管日益严格的背景下,自动抓取并整合多渠道通信内容(如邮件、WhatsApp)的模式,将持续面临合规性质询。

总体而言,folk Assistants代表了一种正确的演进方向:让工具真正为人分担工作,而不仅仅是提供更多功能。它的成功与否,将不取决于AI技术的堆砌,而取决于其能否在复杂的、非标准化的销售实战中,展现出接近人类销售助理的上下文理解与决策智慧。这是一条艰难但值得期待的道路。

查看原始信息
folk Assistants
folk assistants proactively take action from your CRM. They research companies, send outreach emails, recap ongoing deals, and identify who to follow up with.
Hi Product Hunt Community, Simo here, the founder & CEO of folk CRM. When we started folk, we were frustrated by how manual and clunky CRMs felt. Legacy tools require you to do all the work: heavy set up, manually updating every record, painful maintenance. CRMs are static and dumb, waiting for users to update them. We wanted something different. A CRM that feels simple and beautifully designed. Connected to your world & capturing the context behind every relationship. And using this context to automate the repetitive work. A CRM that feels proactive. A CRM that works *for* you. Like having a team of assistants. With folk assistants, you have a team working for you in the background - The Research assistant enriches new companies added to the system, and drafts a tailored research note with data from Perplexity. - The Workflow assistant takes action when certain triggers happens, like sending an email, customizing it with the data that lives in your CRM. - The Recap assistant shares a brief on every person, company or deal, summarizing notes and interactions so you can get up to speed on any relationship. - The Follow-up assistants scans all your emails & WhatsApp. Identifies people you need to follow-up with. Drafts follow-up messages for you, in your tone. Assistants follow you through the full relationship lifecycle, from initial research, personalized outreach, getting up to speed and following up timely. Your folk CRM is proactively doing work for you, so you can focus on relationships. Excited to release this to 4000 client businesses, hear what you think, and as always eager to get your thoughts and feedback!
22
回复

@simolemhandez Congrats on the launch, whats the potential of assigning agents to unblocking deal flow?

0
回复

@simolemhandez Weldone Simo, this is huge.

0
回复

@simolemhandez Congrats guys!

0
回复

Hey Product Hunt,

I’m Jean-Yves, CTO and cofounder.

At folk, we believe the way we work is fundamentally shifting. We used to measure productivity by the amount of tasks a human could complete.

With AI, those limits are being pushed. As an engineer, I can do many more things than I used to: code better, debug faster, brainstorm/research deeper. This is all made possible with tools like Claude Code, Cursor and of course our favourite ChatGPT.

What we see happening in engineering, we bring to the CRM world. Our goal is to provide you with a team of assistants; they do the work you would love to delegate: researching and enriching companies, sending outreach, identifying follow-ups, generating recaps of ongoing deals, and much more.

The best part is that this fits naturally into the existing folk experience. We don’t believe AI should be an afterthought—a chatbot artificially layered on top of your product. It should be deeply embedded, ready to use yet customisable.

And because the preliminary work has already been taken care of, you can work smarter and be more focused on what truly matters !

We’re excited to bring this shift to the world of CRM and look forward to hearing your thoughts.

8
回复

Amazing work! Congrats @simolemhandez , @tiboel and the whole Fold team 👏

7
回复

@tiboel  @adrienvdb thanks for the support!

0
回复

Been following Folk’s journey for a few years now, and this launch is a big step forward. Love the direction the product is taking, congrats to the team and good luck with the launch! 🚀🔥

6
回复

@seantiffonnet thank for the words, much appreciated!

2
回复

I'm addicted to your agents !

3
回复

@mvaxelaire wouhouuu ❤️‍🔥

0
回复

Love it! Congrats guys!!

3
回复

@tombenattar thanks ❤️‍🔥

0
回复
Most CRMs are shipping copilots that can summarize or answer questions—what are the concrete tasks folk Assistants can reliably do end-to-end (e.g., detect follow-ups, enrich, draft, trigger workflows) that others typically fail at, and what would make a team switch rather than just add another AI layer on top of their existing CRM?
1
回复

@curiouskitty Great question. Most CRMs add AI as a copilot that summarizes or answers questions, but still relies on the user to take action.

folk Assistants are designed to work end-to-end. They can:

  • Detect when a follow-up is needed based on real interaction signals

  • Enrich contacts automatically before meetings

  • Draft personalized messages using full relationship context (emails, WhatsApp, notes)

  • Trigger workflows (send emails, create reminders, update pipeline stages) without manual steps

Teams switch to folk because AI isn’t an extra layer on top of their CRM — it’s embedded into how relationships and workflows actually run. Instead of managing tools and prompts, the system actively does the work for you 🙌🏼

0
回复

Capturing context automatically instead of relying on manual updates feels like the right shift for CRMs.

Curious how folk decides what context is important enough to act on versus just store?

1
回复

@shreya_chaurasia19 Thank you for your question 🙂
folk automatically captures rich context from past interactions across email, LinkedIn, calendar, and WhatsApp, along with your own notes — including call transcripts. We then use this context to identify meaningful signals and decide which ones are actionable, so we surface the most relevant suggestions rather than just storing information.

0
回复

Congrats on the relentless innovation and problem solving !

1
回复

@vicgrss thanks for the support! ❤️‍🔥

0
回复
Congrats on the launch 👍 nice app. Automation is the future. However in the planned features that are labeled coming soon, I believe you could hint the users what features are coming. Not totally revealing the feature, just a hint. Just upvoted
1
回复

@zahran_dabbagh thank you Zahran. Indeed, that could be a great option (because we've a lot to come haha!)

1
回复
@simolemhandez wish you all luck ❤️
0
回复

Wow, folk looks incredible! The proactive assistant features are a game changer. How well does it integrate with LinkedIn Sales Navigator for finding new leads? Super excited to try this out!

0
回复
Congrats on the launch 🚀
0
回复

As someone who forgets to follow up 90% of the time… I feel personally attacked (in a good way). Take my upvote :)

0
回复
#4
AI Test Engineer by BlinqIO
Playwright code + self-healing tests w/ no vendor lock-in
245
一句话介绍:一款基于AI的端到端测试工程师工具,通过生成真实Playwright代码并实现自愈测试,解决了软件发布周期中测试自动化维护成本高、易脆弱的痛点。
Productivity Developer Tools Artificial Intelligence
AI测试自动化 端到端测试 Playwright 自愈测试 无供应商锁定 CI/CD集成 代码生成 质量保障 软件发布 智能修复
用户评论摘要:用户反馈积极,认可其节省时间和解决维护痛点的价值。主要问题集中在:AI如何遵循内部编码规范、测试有效性验证、测试优先级推荐、现有脚本迁移、元素完全移除的处理逻辑,以及在复杂场景下与纯手写代码相比的决胜优势。
AI 锐评

BlinqIO的“AI测试工程师”并非又一个华而不实的录制回放工具,其核心价值在于试图用工程化思维解决自动化测试的“最后一公里”问题——持续维护。它聪明地选择了生成真实、可拥有的Playwright代码作为技术基底,这直接击中了企业对于“供应商锁定”的深层恐惧,赋予了产品良好的工程亲和力和退出自由度。

然而,其宣称的“Vibe Testing”和“自主AI能力”需要冷静审视。AI在定位器失效时的自愈能力固然亮眼,但这本质上仍属于模式匹配和上下文理解的范畴,对于涉及复杂业务逻辑变更或全新交互流程的测试失败,AI能否真正理解“意图”并做出正确修正,仍需大量实践验证。评论中关于“测试有效性验证”和“优先级推荐”的提问,恰恰暴露了当前产品在测试策略高阶智能上的空白。

真正的挑战在于,它将测试用例的生成和维护从工程师肩上卸下,但也可能将理解系统行为的责任从工程师眼前移开。当测试完全由AI创建和修复时,如何确保这些测试依然是对系统有价值的、而不仅仅是“能通过”的检查点?这要求产品必须提供远超代码变更的、可解释的洞察报告。如果只能做到“更快地生成和修复测试”,那它只是一个效率工具;若能演进为“基于对应用变更的深度理解,智能地建议测试范围和重点”,它才有机会成为重塑测试范式的关键角色。其即将推出的Jira集成功能,或许是迈向这个方向的一步。

查看原始信息
AI Test Engineer by BlinqIO
BlinqIO’s AI Test Engineer builds and maintains Playwright automation in minutes - generating real code in your repo, integrating with CI/CD, and self-healing when your app changes. No lock-in, just faster, stable, zero-touch testing.

Hi Product Hunt,
I’m Tal, founder and CEO of BlinqIO - the AI Test Engineer for end-to-end testing.

We’re engineers ourselves, building for the technical community, and we’ve felt this pain firsthand:
test automation is one of the biggest bottlenecks in software release cycles - especially E2E testing.

We built BlinqIO to remove that bottleneck entirely.

BlinqIO introduces Vibe Testing:
An AI Test Engineer that uses GenAI to create, maintain, execute, and analyze end-to-end tests automatically.

Under the hood, BlinqIO generates real Playwright test code, stored in your own Git repository - so there’s no vendor lock-in. You fully own the tests.

The result: E2E testing goes from a constant human burden to an autonomous AI capability, finally unblocking faster, safer releases.

It is free to get started.

What’s next: we’re already working on extending the AI Test Engineer to support API testing.

We’d love your feedback: What other testing areas or workflows would you like us to support next with the AI Test Engineer?

Tal



24
回复

@barmeir_tal sounds great, is it possible for the AI to follow our internal coding standards or formatting rules?

0
回复

@barmeir_tal Hi Tal, this sounds very interesting. congrats on the launch. How do you address test planning with this; bugs vs bad test?

5
回复

@barmeir_tal Tal ! quick question: When BlinqIO generates or updates a Playwright test that later flakes or breaks, how does a human quickly understand why that test exists and whether it’s still valid, without digging through AI-generated code?

0
回复

Tests that write themselves? Hold my coffee

Hey everyone 

I am Liat, Head of Marketing at BlinqIO. For the last year our team has been building something called the AI Test Engineer. It started as a small idea and turned into months of talking with testers, breaking things, fixing things, and trying to figure out what actually creates value in real QA teams.

The main challenge was finding the right product market fit in a space that moves fast and has strong opinions. We did not want to add another shiny tool that creates more work. We wanted something that feels useful on day one, respects QA as a profession, and removes some of the pain around automation, coverage, and maintenance.

Along the way we learned what matters most to the people using the product. Speed without chaos. Ownership without lock in. Accuracy without extra complexity. Those values ended up steering the entire roadmap.

Today we are bringing the AI Test Engineer to Product Hunt to get honest feedback from a community that cares about building real products. If you test, ship, or break software for a living, I would love to hear what you think and what you would improve.

Thanks for reading and for being such a supportive community 


10
回复

Hi everyone, I’m Rebecca from BlinqIO.
I’ve been part of the team preparing today’s launch and working with the team developing the NEW version of the AI Test Engineer. A big part of this release is Vibe Testing starting with intent and making automation easier to adopt in real teams. We’re excited to share it here and hear your feedback!!
Congrats team @sapnesh_naik @liat_niederman @barmeir_tal

8
回复

Congrats on the launch! I’ve been using BlinqIO and it’s been impressive. It intelligently updates test scenarios when the UI changes. It’s a real time-saver for creating and maintaining automated tests and speeding up QA. Vibe testing is here 😎

8
回复

@max_feels Thank you Max for the feedback. Vibe testing is defenetly here

3
回复

Hi all!


I’m Sapnesh, Solutions Architect at BlinqIO, and I’ve spent most of my career in the test automation space, thinking on how best to tackle the gap between “tests we want” vs “tests we can actually maintain.”


UI test automation breaks not because teams don’t know how to write tests, but because keeping them running over time is expensive, fragile, and mentally draining.

So we built the AI Test Engineer to work the way real engineering teams work.

From a technical standpoint:

  • BlinqIO generates real Playwright code using the AI Recorder

  • The code lives in your Git repo, runs in your CI, and can be edited in VS Code like any other test.

  • AI Recovery fixes your failed tests: understanding intent, mapping UI changes, recovering broken selectors, and analyzing failures

  • No vendor lock-in. You can export the code and run it on its own without BlinqIO

  • Full freedom to edit/change code - you edit/modify the code if you want to. In fact, you can chat with our AI Test Engineer to implement custom code changes!

  • We also provide our own BlinqIO execution cloud where you can run tests in parallel, with AI Recovery, retry failed tests, get failed test videos, Playwright traces, and more! - You don't need a third-party cloud service to run your tests.

We call this Vibe Testing, starting from intent and letting AI handle the repetitive, failure-prone parts of automation while engineers keep ownership and control.

Please share your questions and opinions here, and I’d be happy to answer them!

- Sapnesh

7
回复

Does the AI recommend which tests should be prioritised based on risk or production usage data?

7
回复

@hiya_chaplot1 Thanks for the question. The AI Test Engineer does not prioritize tests based on risk models or production usage data. Today the focus is on generating, maintaining, and executing real Playwright tests, and letting teams decide prioritization based on their own context. Happy to share more if you are curious.

2
回复
@hiya_chaplot1 Hey, can you clarify what you mean by recommend priority for tests Do you want to tag particular set of tests as high priority so you can execute them frequently, or do you want help in prioritizing which failing tests to fix first based on risk
0
回复

@hiya_chaplot1 We also have a new feature coming soon that will let you connect your Jira project to BlinqIO so that the AI Test engineer can suggest test cases for tickets that you select. This will be released in the next two weeks!

4
回复

Congrats on a long-awaited launch, team @barmeir_tal @liat_niederman @rebecca_villalba @sapnesh_naik

Do tests run locally, or do they go through your cloud? Asking from a security perspective.

6
回复

Hi Kate
Both options are supported - local execution and cloud execution - and you choose based on your workflow.@liat_niederman  @rebecca_villalba  @sapnesh_naik  @kate_ramakaieva 

4
回复

@barmeir_tal  @liat_niederman  @rebecca_villalba  @kate_ramakaieva 

You have full flexibility in determining where you want your tests to run.

  1. Locally on your laptop

  2. Recommended - On the BlinqIO Execution Cloud (Support parallel runs, AI Recovery, Failed test retries, and videos/traces for failed tests).

  3. External CICD/Servers - You can run it on your own third party CI-CD like Github Actions, Jenkins, Azure, AWS etc or any Windows/Mac/Linux server enironment.

Hope this helps!

Sapnesh

5
回复

Nice idea!

5
回复

@sneas Thank you Dimah!

1
回复
👍
5
回复

Congrats on launching!! Can teams realistically go from zero tests to CI-ready suites in under 2 minutes?

3
回复
@roopreddy It takes on average 10 minutes for our customers to create a an end to end test. These tests are usually 15-25 steps long which if you were to code manually would have you at least couple hours! If you have a very simple test with under 5 steps the yes, you could very well be done in 2 minutes!
1
回复
If I already have Playwright scripts, can I upload them and let your AI improve or extend them?
3
回复

@1mirul If your Playwright scripts are in JavaScript, then yes! There is a migration path to import them into BlinqIO and extend them.

3
回复
What happens if an element is completely removed from the page? Does the AI suggest a different action or ask for input?
3
回复

@priyankamandal , it depends on the test step name generated during test creation.

Let me take an example:

Original test step:

User accepts terms and conditions and logs in with "<username>" and "<password>"

Actions: Check the terms checkbox, enter username, enter password, click Login

In the above case, if the:

  • Accept terms checkbox was fully removed: AI is intelligent enough to understand it's no longer required for the step scope and removes the line of code, and continues logging in as usual

  • Login button was renamed to Sign in: AI understands it's a name change and updates the locators accordingly and logs in

  • A new mandatory input was added, example: CustomerID: Since the input is mandatory, the AI cannot skip it, but it also doesn't have enough info to fill the input. In this case, the AI Recovery will fail, and the failure reason will be highlighted as insufficient data to fill the CustomerID input

As you can see, the AI Test Engineer is intelligent enough to understand the scope of the step and recover the failure accordingly!

3
回复
If a team is choosing between BlinqIO, a recorder-based tool, and “just write Playwright with better test IDs,” where do you win decisively?
3
回复

@curiouskitty Great question! I'll answer it using bullet points:

  • Huge time savings during test creation: The AI Recorder generates a full-blown Playwright project with test data, code, and CucumberJS test feature files within minutes.

    If you consider a typical end-to-end scenario with around 20 steps, a human Engineer might spend 1-2 hours coding the test, running it, testing it against different inputs, and then pushing the code. And while doing this, the human engineer might use only one or two locators per action.

    With BlinqIO, all of this takes an average of under 15 minutes!

  • AI Recovery: When you use the AI Recorder, the AI Test Engineer is learning about your application as it captures screenshots, remembers HTML DOM content, and the network calls your app makes. All of this helps the AI perform autonomous recovery of your failed tests. All you need to do is log in in the morning, review the screenshot-based fix report, and accept the AI fix.

  • Open source & No Vendor Lock In: The code is generated using open-source libraries like Playwright, JavaScript, and CucumberJS. Even if you stop using BlinqIO, you can still keep the code and run it on your own (npm run test!).

Hope this answers your question. Let me know!

3
回复

E2E tests being the release bottleneck is painfully accurate. Most teams dread touching them.

Curious how BlinqIO handles flaky tests over time. Is that mostly learned behavior or rule-based?

3
回复

@shreya_chaurasia19 You're right! The BlinqIO AI Test Engineer learns from successful test runs and uses that knowledge to recover failed tests.

Each successful run generates a run report that includes screenshots, console logs, network logs, and the Playwright trace file. The AI Test Engineer analyses this and builds a mind map of your application. It learns about your application just like a human engineer would! And this helps it improve over time at recovering from failed tests.

Attaching an image below on how the AI Mind map looks after you have a few tests automated and run.

3
回复

What happens when Playwright locators change... does self-healing really handle it automatically?

2
回复
@nuseir_yassin1 yes! The AI Recovery analyses the root cause of the failure and if it detects that the failure was due to a UI change, it tries to generate new Playwright code with updated locators to fix the failure. We’ve tested it working on different use cases and websites!
2
回复

From a QA workflow perspective, how does your vision about Vibe Testing change collaboration between developers, QA engineers and product managers over time?

2
回复

@zerotox Good question! I think that, with Vibe testing, test maintenance, and deciding what to test/when to test/which tests to run, all of these critical tasks get much easier as AI becomes increasingly impactful. We at BlinqIO think the QA role will evolve into a supervisor role for multiple AI agents that do most of the hands-on work!

1
回复

Do you see AI eventually discovering test cases automatically based on how users interact with the product?

2
回复
@krishna_gupta51 Yes! This is part of our roadmap. We also  have a new feature coming in that will connect to your Jira and generate test cases from your Jira tickets.
2
回复

@liat_niederman Congratulations. And happy product launch.

1
回复
0
回复

Wow, this looks amazing! BlinqIOs AI Test Engineer building Playwright code directly in my repo is a game changer. How well does the self-healing adapt to more complex UI changes, like refactoring a whole component?

1
回复

@jaydev13 Thanks! The BlinqIO AI Test Engineer is extremely good at even complex UI changes that cause test failures!

One customer had a test that filled out a large transport order form with multiple sections in a single long page. Later, their product team refactored that flow into a multi-step form wizard, splitting each section into its own step.

And when the test failed, the AI Test Engineer understood that the test scope remained the same even if the UI structure had changed. The AI Recovery automatically generated new Playwright code and locators for the UI and recovered the test without any manual intervention!

1
回复

Can you still tweak the generated code or is it a black box? BTW, you're top 5 on the leaderboard today :D

1
回复

@himani_sah1 Thanks! The AI Test Engineer commits the code to your own Git repository (GitHub/BitBucket/GitLab), so you have full access to the code! You can tweak or change this code easily right within your BlinqIO project, and you can even chat with the AI Test Engineer to make the code changes for you!

Our customers use the AI Code chat to implement complex steps like

  • Connecting to a DB, getting data, and using that to fill inputs in the UI or assertions.

  • Opening Excel files, reading them, and using them in tests.

  • Generating test data with a specific logic

  • and more!

2
回复
#5
Kiki for Mac
The accountability monster that keeps you focused on a task
236
一句话介绍:一款通过虚拟“问责怪兽”角色和趣味性干扰惩罚机制,帮助Mac用户在需要深度专注的任务场景中克服分心、保持任务执行力的桌面应用。
Productivity Task Management Time Tracking
生产力工具 专注辅助 行为矫正 趣味交互 Mac应用 虚拟伙伴 应用屏蔽 防分心 问责机制 创意工具
用户评论摘要:用户普遍赞赏其趣味性和有效性,认为其巧妙结合了问责与乐趣。主要问题集中在付费模式(期待一次性买断)、功能细节(如网站屏蔽粒度、与番茄钟整合)以及多平台支持(Windows/Android)。开发者回应将增加一次性购买选项。
AI 锐评

Kiki for Mac 表面上是一款以“可爱怪兽”为噱头的专注力工具,但其内核揭示了一个更深层的用户痛点:传统“阻断式”工具在知识工作时代的失效。当社交媒体等显性干扰被屏蔽后,用户的分心行为已进化为在“看似productive”的工具(如Notion、Slack)间进行无意义切换,这是一种更具欺骗性的拖延。Kiki的聪明之处在于,它放弃了“堵”的思路,转而采用“情境定义”与“行为反馈”。

其真正价值在于两点:一是**意图仪式化**。通过让用户手动选择“本次专注所需的应用”,它将模糊的“开始工作”转化为一次清晰的情境声明,强化了心理承诺。二是**拟人化负反馈**。Kiki的“生气”、“打字”、“搞破坏”等行为,将抽象的时间浪费转化为具象、有趣且略带压力的情感互动。这种设计利用了人类对社会评价和情感联结的本能反应,其惩戒效果比冰冷的计时器或封锁页面更为深刻。

然而,其挑战也同样明显。首先,趣味性可能导致“ novelty effect”(新奇效应)快速消退,一旦用户对Kiki的互动脱敏,工具效力可能骤降。其次,其商业模式依赖订阅制,这与“运行于本地”的特性存在认知冲突,是用户评论中主要的质疑点。长远看,Kiki更像一个精巧的行为实验,它验证了“情感化问责”在生产力领域的潜力,但其可持续性取决于能否从“有趣的玩具”演进为能适应专注力疲劳周期的“智能伙伴”。它抓住了后阻断时代专注工具的空白,但最终是成为昙花一现的网红应用,还是开启人机协作专注的新范式,取决于其后续能否在行为科学模型上构建更深的护城河。

查看原始信息
Kiki for Mac
look, we both know how this goes. you sit down to work. you check one thing. then it's been 40 minutes and you're watching a guy restore a rusty knife on youtube. 〰〰 kiki lives on your mac and helps you actually finish the things you start. choose your focus. pick your apps. hit start. kiki works alongside you, typing away in the corner. 〰〰 get distracted and kiki has opinions. flips your lights. sets things on fire. focus and you might just finish that project you've been avoiding.

Hey Product Hunt, I'm Isaac.

I made Kiki because I'm distractible. Like, impressively so.

I already use blocking apps. Instagram is gone. Twitter is gone. The obvious stuff is handled. But somehow I'd still sit down to write and end up reorganizing my Notion for two hours. Technically productive. Completely off task.

Kiki isn't about blocking. It's about intention. Pick one thing. Choose the apps you need. Everything else—even the "productive" stuff—is a distraction.

Wander off and Kiki gets pissed. Starts with a nudge. Then Kiki's flipping your lights on and off. Setting things on fire. Keeping you on track.

It worked for me. Then friends started asking for their own. Now Kiki's here for everyone.

If your problem isn't just social media but "let me quickly check Slack while I'm supposed to be writing"—this might be for you.

Kiki's giving away 20 free annual subscriptions to the best (or most desperate) focus tricks in the comments (whether it worked or not). Kiki will judge.

For everyone else, we're offering 20% off to celebrate our launch with code PH20.

Oh, and if you're curious: I'm a designer—spent a long time at Google Creative Lab, started a studio called Early Works, and have shipped things like Google Recorder, Asterix, and Moon Mirror (we turned a solar facility into the world's largest moonlight-powered display). I also used Kiki to make Kiki. So, you know. It works.

14
回复

@blankensmith Very cool. congrats on the launch. Rusty knife ftw.

1
回复

@blankensmith As a fellow side-quest specialist I need this! 🏃🏃🏃

My current focus hack is noise-cancelling headphones + Steve Reich "Music for 18 musicians" — gets me in the ZONE.

0
回复

Congrats on the launch! Kiki’s “focus or else” vibe plus app-level blocking feels like the perfect mix of accountability and fun for getting deep work done on Mac.​

4
回复

@zeiki_yu Thanks! We really tried to balance making an app that helps us get serious work done, without taking itself too seriously

0
回复

I love Kiki! What a fun way to slap myself around to keep me on task! 😂💥

3
回复

This is the kind of whimsy + usefulness I miss from the internet. Such a clever and practical tool. Will become a mainstay on my desktop for years to come.

3
回复

Congrats on the launch! I'd love to use it! Question: it runs offline/locally, doesn't it? Why don't you guys offer a pay-once/lifetime purchase option as well?

3
回复

@nzsalamon Hey Nestor, thanks! Yep, it runs locally and works offline. You're one step ahead of us. We're planning to add a one-time purchase option soon. Hope you give Kiki a try!

1
回复

OMG this is amazing! I love the idea of Kiki typing away and roasting me when I get distracted. Does it integrate with Pomodoro timers for even deeper focus sessions?

2
回复

@jaydev13 Yep! It has a built in pomodoro timer.

1
回复

Awesome. Can't wait to try it out. This is made for me.

2
回复

Ahh I love this. The illustrations and UI are so fun.

2
回复

I will not let Kiki down!!! Nice branch / product / launch.

2
回复

Congrats on the launch!! Love the design and can't wait to use it to be more productive :D

2
回复

Congrats! Does this only block apps or also websites?

For example, blocking my email/linkedin on Google Chrome but can still access other websites?

2
回复

@brandon_smith32 You can block specific websites like your example and still use everything else. You can also just allow specific websites, and then the rest of the internet is blocked. We've had people who want to lock themselves into just specific online courses, but not be able to access the rest of the internet.

1
回复

I've been testing Kiki for a few months and she's become my little companion buddy when I do hard, annoying, work that makes me easily distracted. Especially for writing and thinking tasks, Kiki helps me avoid seemingly "productive" work that's really just busy-work keeping me from doing important things.

2
回复

i love kiki. when does the merch drop?!

2
回复

exactly the kind of chaotic-good energy I need to stop me from ending up three hours deep in a rabbit hole. Love the character and the 'tough love' approach to focus! Huge congrats on the launch, this is adorable!

2
回复

Such a great idea, and I love the little monster! I can't wait to try it when windows or android comes out, but for now I'll have to ... hold on ... just got a message ... sorry, just let me finish this email ... where was I? Oh I'm sure it wasn't important.

2
回复

I’ve been testing out Kiki for a couple of months and I love it. It makes a lot of other software feel overly serious, but doesn’t fall into the Clippy trap. I find the playfulness also breaks the dogmatism of these kinds of tools and—as someone who works on their own—it’s strangely comforting to watch it working alongside me.

2
回复

Early Kiki user here. I’m usually pretty good about staying on task, but on hectic days when I need to lock in on deep work (e.g. a budget or a strategy doc) Kiki’s become my go-to work buddy. Would love it if there were a few more “angry” animations besides fire and the light switch. Also, nice bonus I didn't expect – my kids love Kiki!

2
回复

I was lucky to be an early tester of Kiki. I’m also highly distractible and have used A LOT of productivity apps (one might say I’m distracted by them 😃).

What I love about Kiki is the irreverent delight. Lighting my screen on fire when I pop over to check the news even though I have a deadline and need to finish the deck I'm working on? I don't know the psychology of it, but a little laughter when my focus failure is exposed helps me get back on task. I can't wait to see what surprises the Kiki team has in future updates.

2
回复
Congrats on the launch. A smart idea and I loved the idea of not randomly blocking apps but doing that for a specific task 🚀
2
回复

@zahran_dabbagh Thanks! That kind of task based work is exactly how i need to focus to get things done...like working on making Kiki!

0
回复

This is such a fun product — I love the creativity and personality in Kiki!

Quick question: do you plan to support Windows and mobile devices in the future?

Sometimes my biggest distractions come from my phone, so it would be amazing to have Kiki keeping me on track across all my devices.

2
回复

@mrpop Hey Alex! We're playing around with mobile right now. The goal is to help people focus while they're on their computer, but yeah, phones are often the thing pulling us away.

We're exploring a simple Kiki app that just locks your phone when you start a session on your computer. No promises, but if testers tell us it's worth it, we'll ship it.

Windows too. We'd love to, but we're a small team. If we hear from enough people wanting it, we'll build it.

1
回复

Congrats on the launch!!! Can’t wait to try it out!

1
回复

incredible product because its incredibly simple but it WORKS. everyone working on a computer struggles with some degree of distraction.. its the nature of digital work; so much of our life is in one place. Kiki removes the noise and makes your computer a place of actual productivity.. when you want it to be! brilliant.

1
回复

I'm a designer and this has been incredible for long work sessions. Usually I start designing and drift into Slack, inspiration boards.. Kiki just says nope, you picked Figma, stay in Figma!

1
回复

Have had the chance to test and use this for a few months

I’m a lifelong focus seeker (often a failure). I have physical world hacks, a hexagonal clock that splits hours a bunch of different ways, the way I arrange my desk, even how I arrange myself near portals of distraction (aka windows). 

I use kiki every day, it works 2 ways for me, mindfulness, when I’m ready to get to it, it helps (forces me, it a good way) to think about what I need. And then it pulls me back. I always slide, and kiki keeps me on it for 20 min or an hour or..... . 

Kiki isn't subtle, which is amazing.

One more thing. You will get immediate action from the Kiki team, incredibly responsive to feedback.

1
回复

Are there plans to add project tags or a task list that you can work towards?

I’ve been using Kiki over the past couple months. It’s been been great for tracking my time and seeing where my hours are going. Excited to keep using Kiki and see how it evolves!

1
回复
Add a pomodoro timer!
0
回复

@anthony123345 Kiki has a built-in pomodoro timer! You can choose how long you want to focus for and it'll automatically remind you to take a break between sessions.

1
回复
People already use combinations like Pomodoro + separate blockers, or strict tools like SelfControl. What’s the clearest “switching reason” you’ve seen in real use—i.e., what does Kiki consistently do better than timers, blockers, or OS-level Focus modes?
0
回复

@curiouskitty Hi friendly robot. Kiki isn't trying to out-block blockers or out-time timers. Kiki makes focusing feel like something you're doing with someone, not something being done to you. You pick one thing. Kiki joins you. Wander off and Kiki reacts. It's accountability meets playfulness. That's the switching reason we hear most, people actually enjoy using it.

1
回复
#6
Indy
AI support for your ADHD, not just your to-do list
169
一句话介绍:Indy是一款免费AI驱动的ADHD支持应用,通过未来映射、每日脚手架和引导反思等结构化流程,帮助ADHD人群在日常生活场景中建立可持续的执行系统,解决计划与行动脱节的痛点。
Health & Fitness Productivity Health
ADHD支持 心理健康 AI健康助手 行为科学 数字疗法 生活规划 认知辅助 免费应用 结构化提示 未来映射
用户评论摘要:用户肯定其科学基础和人性化设计,认为它超越了传统待办清单。核心反馈集中在:如何避免成为制造愧疚感的工具、如何设计温和的推动机制,以及赞赏其关注长期福祉而非单纯生产力。团队详细回应了设计理念。
AI 锐评

Indy的亮相,与其说是一款新应用,不如说是对当前“AI+健康”赛道泛滥的通用聊天机器人模式的一次精准批判。其真正价值不在于“用了AI”,而在于“刻意限制AI”——用结构化流程取代开放对话,用短期交互对抗认知过载。这直击了ADHD支持乃至整个数字心理健康领域的核心矛盾:更多的功能和信息流,往往意味着更大的用户负担。

产品将“未来映射”置于核心,是颇具胆识的策略。它试图解决ADHD人群的根本挑战:时间感知障碍(难以连接现在与未来)和动机缺失。通过构建个人叙事和长远愿景来驱动当下行动,这比简单分解任务更接近ADHD教练的本质。然而,这也将其置于最大风险区:这种高度抽象、依赖内在驱动的模式,如何量化效果、维持用户长期参与?团队回避“连续打卡”机制是明智的,但“14天挑战”是否只是另一种变体,仍需观察。

从评论看,团队对“避免制造愧疚感”有清醒认知,这是产品伦理的关键。但其商业模式隐忧已现:“免费”模式如何持续?未来是否会对高级功能或数据分析收费?一旦涉及付费,其倡导的“无压力”环境是否会变质?此外,其方法论高度依赖创始团队的教练经验与科学架构,规模化过程中,AI的“适应性系统”能否保持干预的个性化与精准度,将是严峻考验。

Indy代表了一种可喜的转向:从“用AI管理疾病”到“用AI赋能神经多样性大脑”。它的成功与否,不仅关乎商业前景,更关乎我们能否用技术构建出真正尊重人类复杂性的支持系统。

查看原始信息
Indy
Indy is a free, AI-powered ADHD support app designed for how ADHD brains actually work. Informed by behavioral science and insights from 80,000+ coaching sessions, Indy uses adaptive systems and structured prompts to help you stay connected to what matters through future-mapping, daily scaffolding, and guided reflection.
Thanks @garrytan for hunting us! Hi Product Hunt! 👋 I’m Chris, co-founder & CEO of Shimmer (we build ADHD support). Today we’re launching Indy. It’s a free AI-powered ADHD support app built from what we learned across 80,000+ coaching sessions, plus our beta community. Quick background: I was diagnosed with ADHD in 2022. I got the “here’s meds, come back if you need more” experience. What I actually needed was structure & strategies that worked in real life. I’ve now been in ADHD coaching myself for 3 years, and it’s been transformative. But still, coaching isn’t accessible to everyone, and even when you have it, the hard part is executing in the space between sessions. That gap is what Indy is for. Indy is not a generic chatbot and it’s not a to-do list. It’s a purpose-built AI system designed around how ADHD brains plan, remember, and follow through. What Indy does: - Future mapping (Lifeline): map key moments from your past, present, and future so what matters feels real again - Daily + weekly check-ins: lightweight intentions + reflection that reduce friction without turning into “productivity pressure” - Insights over time: Indy tracks wins, effort, and patterns so progress doesn’t disappear - Problem-solving when stuck: science-backed frameworks to figure out what’s blocking you and pick the smallest next step Indy is free, and we’d love your feedback. If you try it, tell us what feels surprisingly helpful and what still feels hard. Chris Wang Co-founder & CEO, Shimmer
17
回复

@christal congrats on the launch!

0
回复

@garrytan  @christal Congrats on the launch Christal. How (practically) do you handle gentle nudgebacks into task flow and keep that compounding?

2
回复

@christal To me, the essence of entrepreneurship lies in identifying problems from our own experiences and turning them into solutions. I find this product highly meaningful because it transforms a real solution into a scalable product that can help many more people. I will definitely download it and give my support.

0
回复

Thank you so much @garrytan for hunting us!

Hi everyone! I’m Vikram, co-founder & CTO of Shimmer. 😊

We’re excited to launch Indy, a free AI-powered ADHD support app.

From an engineering perspective, we obsessed over one question: how do you make AI useful for those with ADHD without turning it into an endless chatbot or a noisy productivity tool?

Indy is intentionally the opposite of most AI products:

  • short interactions, not infinite conversations

  • structured flows, not blank-page prompting

  • designed to reduce cognitive load, not create more

  • built to be helpful in real moments, not just give “good answers”

We built Indy with our beta community and iterated fast based on real usage. If you have feedback at all whether positive or negative, please share it with us at support@shimmer.care and we’ll do our best to respond quickly so that we can improve the app based on your feedback.

Thank you PH community for supporting our launch today and excited to hear what y’all think!

Cheers,

Vikram – Co-founder & CTO, Shimmer

8
回复

Hey everyone! I’m Anthony, one of the engineers at Shimmer. We’ve had a ton of fun building Indy, and honestly it’s been really cool seeing the impact it has already made on people's lives.

From the start, we wanted to be super intentional about structure. The goal wasn’t to make something that talks forever, it was to make something lightweight that helps you get what you need in terms of structure and clarity, so you can go and tackle whatever is most important for you in your life.

If you’re curious, I’d love for you to check it out. And if you have questions, feedback, or ideas, feel free to leave a comment!

7
回复

@anthony_lin It's crazy to see all your hard work come to life man! Been an exciting journey with Indy so far and can't wait to see the next stop!

3
回复

Thank you so much @garrytan for hunting us! Hi everyone! 👋

I am Xenia – the Principal Scientific Officer at Shimmer.

As a person responsible for evidence-based approach to Shimmer products, I am very excited about this launch.

Most of ADHD apps treat human like a machine who needs to be “trained” from getting distracted to fixing their eyeballs on the goal for long enough to receive a pat on the back.

Others attempt to reduce human complexity into what can be measured and averaged, at the expense of nurturing a singular subjective experience of each person.

These approaches ignore the most important thing: humans are not machines. We are living our life story as we go along, making meanings in interaction with our context and other people.

Altogether, this results in a never-ending loop of responding to what feels urgent…

Indy is radically different.

The modern developmental science suggests that we should instead focus on what is important first. What do you want to see happening in your life this year? In five years? Thirty years on?

Else, existential psychology shows that despite the uncertainty of the modern world, there is much more choice and freedom available to us. There is every reason to invest into the strategic life-span view - and this is what Indy does.

I would love to answer any questions you might have and look forward to you trying it out!

 

7
回复

@xeniaangevin Kudos to all the work you put in to set the foundation for Indy! We couldn't have done this without you and it's exciting to see how helpful Indy has been already from our early set of users!

1
回复

@xeniaangevin It's been such a pleasure working on Shimmer and Indy with you! We wouldn't be where we are without the level of insight and innovation you bring into every interaction and project we have! Cheers to Shimmer, Indy, and the future of ADHD!

0
回复

This is soooo iconic huge congrats on the launch !!!! Sharing with my friends now

5
回复

@anthony_scarpone_lambert Thanks so much for the support and really appreciate the shares too!!

0
回复

Congrats on the launch! Love how Indy pairs behavioral science with adaptive AI to support ADHD brains beyond checklists, focusing on meaningful momentum and self-trust instead of pure productivity.​

5
回复

@zeiki_yu Thanks so much for the support! Let us know if you have any feedback if you happen to try it out too :)

0
回复

Finally, an ADHD support app that’s practical and science-based. Structured prompts and future-mapping could make a real difference for people who struggle with traditional tools.

5
回复

@leotrim_lota Yes! We've heard so much from our beta users how other apps are not structured well for ADHDers. They do things like nudge you on to-do list items or break down tasks, but don't help with the deeper parts of ADHD: motivation over the long term, task paralysis, problem solving and solutioning in a way that's personalized to your unique ADHD, the list goes on! Thanks for the kind words!

2
回复

Hey Product Hunt Community!

I’m Alex, Shimmer’s Senior Manager of Care & Coaching and I’m super excited that the launching Indy expands the ways in which we can serve the ADHD Community!

We know there’s so many great forms of service out there, but it's often inaccessible due to cost or time constraints, and the current apps on the market really don’t focus on long-lerm, future-orientated wellbeing for the ADHD Community. I am so honored to get to be part of an initiative that focuses on making care more feasible.

What I love about the space we created is that it creates a consistent space to pause, reflect, and reset. It gives people a way to reconnect with what matters, which can be so challenging to do.

I’m excited to see how people use it, and what we learn from the community next.

Best,

Alex

4
回复

@alex_bellitter Have to give a huge shoutout to you for all the care and intention you put behind all of the guided exercises and flows within Indy! It's amazing to see all the work you were drafting on Notion docs and figjams to finally be out in the real world!

1
回复

We need more apps like this! Thanks from a fellow ADHDer 🙋🏻‍♀️

4
回复

@berins Thanks so much Berin and really appreciate your support for Indy!!

0
回复

Hi, everyone! I’m Elisabeth, the Product Designer of the team.

Indy is not just another ADHD tool. There are plenty of productivity lists, calendars, and trackers. The Indy experience is instead rooted in narrative and vision, to help you identify the meaning behind doing any of those things on your lists, calendars, and trackers. This is something you experience on a deep level when working with a professional Shimmer ADHD coach.

To communicate the depth of the experience, we explored a celestial, dark-mode aesthetic with a lot of clean, open space that helps the ADHD mind focus on one thing at a time.

We hope you download and enjoy the app as we continue to evolve the UI and features to deliver more value and delight!

3
回复

@elisabethbridges The celestial theme is SO beautiful, you absolutely crushed it Elisabeth! Can't wait for future features and how we extend this theme!

1
回复
Hi Product Hunt! I’m Trina, and I’m on the Growth team here at Shimmer! 👋 So much care went into building Indy and we’re so excited to share it with you!
3
回复

@trina_haynes Couldn't have scaled our impact without you and making Indy what it is today!

1
回复
ADHD users are especially sensitive to feeling pressured or judged. How did you design Indy’s prompts, pacing, and progress tracking to avoid turning into a guilt machine—while still creating enough accountability to change behavior?
3
回复

@curiouskitty This is something we've thought a lot about and specifically why we don't have streaks. We do have 14-day challenges that allow you to "lock in" when you choose to, but also you won't approach these when you're on vacation or in "low spoons" mode. The goal isn't just productivity or goal achievement, but it's life well-being, so that might mean investing in self care or rest when it makes sense!

2
回复

I am loving Indy so far. Having a guiding theme for my week and then checking in every morning and evening is proving to be the structure. I need to get more done and also notice the wins from the day.

3
回复

@laura_boggess So glad to hear it's been helpful! Definitely let us know if there's anything we can ever do to improve the experience for you as you use Indy!

1
回复

@laura_boggess So glad you're enjoying Indy and getting value from it. Proud of your progress and excited for the future for you!!

0
回复

As someone who also has some ADHD users, this is noticeable.

If you can learn people how to turn this "diagnosis" to skill, you can do miracles.

3
回复

@busmark_w_nika Yes! There are so many under-noticed and under-leveraged strengths that ADHDers have, once they gain the awareness and self-confidence to explore them. We'll actually be building out a structured strengths-finding flow as well later this year!

2
回复

@busmark_w_nika Thanks so much Nika! Your support helps us a ton!

0
回复

Congrats!

2
回复

@kingsongchen Thanks so much Kingsong!!

0
回复

Indy has been very useful in just helping me track some of my learnings about how I work best and identifying blind spots or things I hadn’t considered. It’s become a part of my morning routine!

2
回复

@tiasabs Thanks so much Lavanya! Love that Indy is being integrated that way into your day, and we'll make sure to keep improving it's pattern recognition!

0
回复

I've been using Indy since it became available on Play Store and I love the daily checkins. It's like having that friend I always needed to check in on me, remind me of my values, my goals. This is what I've been trying to accomplish using all the notion templates and journals under the sun.

Truly fantastic job.

2
回复

@elcicneha Thanks so much for the support Neha! Glad you're loving the daily planning/reflection chats, let us know if you ever have any feedback about them as you continue to use Indy! :)

0
回复

Really like the framing of “not just a to-do list”.

Curious what you’ve noticed around when structure helps vs. when it hurts – especially for ADHD.

In my experience, some thoughts need a place to linger before they’re ready to become tasks at all.

2
回复

@dmitry_petrakov Yes, totally and great point! We actually have two input styles - one structured for morning / night, and one for "quick adds". We've found that the usage from our beta users was about 50 / 50 for those two input formats! We're also working on ways to improve memory so that Indy better holds on to things and surfaces them in the right context so you don't need to remember them all but also don't need to open up an overwhelmingly long list!

2
回复

Wow, Indy looks amazing! The focus on future-mapping is exactly what I need. How does the AI personalize the daily scaffolding based on my past reflections? This could be a game changer!

1
回复

@jaydev13 Great question! Our AI tracks what you mention in your reflections and is able to recollect them during your daily planning chats, so it would recall what you were working on before or what you wanted to adjust from the previous day(s) and personalize the support around that rather than just generically helping you plan your day.

0
回复

Indy is game-changing. I have been using it since launch and I am seriously so grateful for it. Came across it through @christal's page and cant thank you enough!

The design is very clean and intuitive. It was easy to get started and to keep using it consistently.

The 'Lifeline' exercise is my favorite feature so far. Having my goals and dreams for the future mapped out in categories really help anchor it and the I love how the app reminds me of these when solving problems or setting daily goals.

The messaging feature was a bit tricky to get used to. Since I am used to AI chat apps letting me ramble. But I am starting to see its strength in helping me stay focused, reduce overwhelm, and explore new possibilities.

Looking forward to see how the app keep generating insights and patterns that would help me in the long term.

1
回复

@sruthy_sasi so glad you found us through Chris's page and thank you so much for giving us a try! We definitely went back and forth between allowing our AI to talk at length or not, but decided that the limitation would help with the structure and push forward a lot of our beta users were needing.

Really hope Indy can continue to help you tackle the day-to-day responsibilities and keep moving forward towards your goals!

0
回复
#7
Binary
Hiring software that is fast, simple and AI powered
157
一句话介绍:一款面向初创及中小团队、AI驱动的轻量级招聘软件,通过简洁高效的操作界面和AI辅助筛选等功能,解决了传统ATS系统臃肿、昂贵且操作繁琐的痛点。
Hiring Productivity
招聘软件 ATS AI招聘 轻量级 初创企业 SaaS 人才筛选 视频面试 用户体验 性价比
用户评论摘要:用户普遍赞扬其设计精美、操作轻快、AI辅助实用,尤其肯定视频问题、LinkedIn预览和收件箱功能。有效反馈包括:明确其适用于中小团队而非大型企业;询问从其他ATS迁移的具体理由;关注是否专精于技术招聘。
AI 锐评

Binary的亮相,精准刺向了传统ATS市场的软肋:功能过剩带来的臃肿与高价。它并非在功能广度上挑战Greenhouse等巨头,而是做了一场果断的“减法”,聚焦于招聘流程中最核心的筛选与沟通环节,并注入AI辅助这一“增效”变量。其真正的价值主张是“够用的智能”——通过视频问题筛选高意向候选人、AI快速初筛、集成化收件箱,这些功能直指初创公司“人手少、申请多、需快速决策”的核心场景。

然而,其“减法”策略也是双刃剑。产品明确将自身定位在“初创与中小团队”,坦承在深度报表等企业级功能上的缺失,这既是市场细分下的清醒,也划定了增长的天花板。评论中“从其他ATS迁移的三大理由”全部围绕“减负”和“成本”,恰恰说明其当前吸引力主要来自对现有产品不满的“逃离者”,而非被其独有创新所吸引的“追寻者”。AI功能被反复提及,但描述多为“好用”、“神奇”,缺乏对其算法边界和公正性探讨,这在AI招聘工具面临日益严格监管的背景下,可能潜藏风险。

总体而言,Binary是一款市场定位极其清晰的产品,在细分赛道内提供了优雅的解决方案。但它面临的挑战同样典型:如何在保持轻量的前提下,满足成长中客户日益复杂的需求;其AI能力的护城河究竟有多深;以及,当巨头们也开始着手“简化”自身产品时,Binary如何保持领先。它更像是一把精心打磨的“手术刀”,锋利但应用场景明确,而非试图成为包罗万象的“瑞士军刀”。

查看原始信息
Binary
Binary is a new-age ATS built for those who want something simple yet powerful. It comes with a free forever plan that supports unlimited jobs and unlimited candidates.

Hi everyone,

We are excited to announce Binary, a new kind of hiring software.

Current hiring tools (ATS) are slow, bloated and expensive. We wanted to build something that's fast, simple and affordable. Binary lets you create exciting job pages, ask video questions, review faster with AI, respond to candidates quickly and much more.

We are grateful to 200+ startups who joined our beta in the past few months and helped us build a product we all love using. Some of these names include Sundial, Supermemory, Lossfunk and Weaver.

Let us know what you think!

27
回复

@shikhars_ Congrats on the launch Shikhars. Whats the one "shut up and take my money" element that you're going for?

1
回复

Great product with beautiful design! All the best 🚀

0
回复

@shikhars_ congrats on the launch! Is it specialized mostly for the tech recruitment?

0
回复

Congratulations on the launch!

I can't imagine hiring without binary. Absolutely gorgeous UI and love the UX and quality of life details + AI assistance is fire :)

2
回复

@shikhars_ Has put in so much care and attention to details in this product. Every single pixel is there to make the best candidate experience!

Anyone who's hiring and want a no fuss and affordable yet powerful ATS, definitely try Binary.

1
回复

Having personally used binary, and also worked on it for a brief period i'll try to be honest about what i think Binary brings to the table :

- Super lightweight and easy to use/setup.
- Tools like Video questions- help me sort high intent users.
- Extremely proactive team helping with any issue, or a feature request.
- Super cheap for product that lets me reply and manage to thousand of applicants.
- Simple to manage candidates between different stages of their recruitment journey.
- Honest Pricing
- Simple to use AI assistant that just seems to work, and its integration with filters is the magic.

I cannot honestly recommend this product more. My most recommended product to people and teams in the last few months.

1
回复

We have been using Binary for a year for all the roles at Mave. Binary makes it super easy to go through hundreds of applications in an hour.
Love the experience :)

1
回复

Have tried it, loved the design and keyboard shortcuts, it makes hiring so much simpler. Good luck!

1
回复

Big fan! I’ve worked on LinkedIn jobs and have hence seen a lot of ATSes. Binary hits the sweet spot for so many small to medium businesses who often don’t want the bloat that comes with the big ones (who have primarily designed their products for large enterprise needs).

1
回复

@raveesh Thanks! You described it better than we did perhaps :)

0
回复

Love the app interface. Great job!

1
回复

We recently got 400+ applications for a role, and Binary made reviewing them a breeze.


The LinkedIn profile preview is brilliant - lets you quickly glance at a candidate's background while reviewing their application. The video input option is simple but effective, especially for roles where language fluency matters.


The team is super responsive too - fixes issues within minutes.


Loved the inbox feature as well: a simple, organised way to communicate with candidates without exposing your email.


Binary sits perfectly between Google Forms and those clunky legacy ATS tools. Exactly what I needed.

Congratulations on the launch, folks!!

1
回复

@devenbhooshan 400+ applications is crazy :) I am glad you loved it.

0
回复

We have been using binary at ChessBase India since early days and love how it has evolved. It finds a sweet spot between a simple google form and a bloated ATS for hiring, doing exactly what is needed.

Replying back to everyone and other AI filters are just cherry on top. Congrats on the launch!

1
回复

@theadxy Thanks Adarsh!

0
回复

That's very interesting, videos are a great way to skip through candidates who simply use chatgpt to submit proposals

0
回复

This is so good. Exactly what I needed.

0
回复
This is an awesome product, just used it for hiring for a product role and was super impressed by it. Highly recommend.
0
回复

@pratyush_r8 Thanks for being an early adopter!

0
回复

Been using Binary since a while now and it's amazing how the product has evolved over time. Kudos to the entire team for building an amazing product 💯

0
回复
If a team is currently on Greenhouse/Lever/Ashby/Workable, what are the top 2–3 concrete reasons they switch to Binary—and what’s the one reason they shouldn’t switch yet?
0
回复

@curiouskitty Thanks for the question!

We've had several teams who moved from their existing ATS to Binary and the top 3 reasons have been:

1. They find their ATS slow and bloated
2. They are bombarded by applications and the AI features are either non-existing or very less useful
3. They feel the price that they're paying for the ATS is way more than the value it is providing

One reason for not switching > If you are an enterprise with dozens or hundreds of open roles, the reporting features in some of the bigger ATS are much more comprehensive than we have today. It's built more for startups and smaller teams instead of enterprises.

0
回复

Congratulations on the launch @shikhars_ I am using Binary as a much smoother interface than Google forms.

The output’s readability has jumped 10x and that makes it for a broader use than as an ATS

0
回复

@sia_steel Thanks! And you are right, we've had many customers using it for scholarships and other use-cases which involve inbound management.

0
回复

Congrats on the launch @shikhars_ . loved the launch video. The designs seems slick

0
回复

We've been using Binary at Creso, and the experience has been amazing. It's super easy to use with a modern, intuitive UX—the platform's look and feel alone make for a great pitch when recruiting candidates to a startup like ours.

0
回复

@agarwalpranay96 Thanks for being an early adopter!

0
回复

If AI replaces anything in the world, I would want it to be HR/ATS systems. Because it's broken. They are playing a bottle spinning game: hidden messages, answers, etc. Not rocket science what HR people do, and it's not efficient at all. And all HR people think the applicant is talking and allocating time only for that particular company, but unfortunately that's not how life works the way they think it does. Glad to see a product in this field. Congrats on the launch!

0
回复

@mehmetkose Thanks and we can't agree more!

A lot of recruiting time is wasted on manual work and we believe AI will play a big role there. Most of our users are able to review/respond to 100% of the candidates reaching out to them within 2-3 days, and we hope more companies can get there soon.

0
回复

Top tier product! Loved how much thought has been given to this!

0
回复

@devansh_pawan1 Thanks for being an early adopter!

0
回复
Amazing 😻
0
回复

@suhasmotwani Thanks for being an early supporter.

0
回复

Good luck!!

0
回复
Congrats on the launch 🚀 i explored a bit and really lived the design of the landing and the clear pricing, which you usually don’t get for major ATS. However, it is not totally clear how it is differentiated from other major ATS and specifically Greenhouse
0
回复

@zahran_dabbagh Thanks Zahran, we've worked hard on the design so I'm glad you liked it!

When it comes to comparison with Greenhouse, the biggest differentiator we believe is simplicity and affordability. Greenhouse is well-suited for large enterprises while we're going after fast-moving startups.

With Binary, you can make your job posting live in a couple of seconds versus the multi-day onboarding process on other ATS which often requires an upfront contract. There are many other details which we've worked hard on - video questions, AI assistant, keyboard shortcuts and more.

1
回复
@shikhars_ That’s amazing and it should resonate with many startups. I believe then you should highlight this in a more prominent way so the added value is captured upon visiting the landing page (just my personal perspective but you know the product better). Wish you all luck and hopefully we’d have a sort of a collab in the future since I have a software that almost operates in the same domain just from a different perspective or from a different side 👍
0
回复

I’ve been using Binary.so for a long time for hiring, especially inbound leads - it’s honestly been solid. The new extension finally makes outbound reachouts easy too. Super excited to try it!

0
回复

@siddharth_bulia2 Thanks for being an early user!

0
回复
#8
LyzrGPT
Private, secure & model-agnostic AI chat for enterprises
150
一句话介绍:LyzrGPT是一款私有化部署、模型无关的企业级AI聊天平台,通过在客户自有生态内部署,解决了安全至上的企业在享受先进AI能力时面临的数据隐私、供应商锁定和合规性难题。
Sales Marketing Artificial Intelligence
企业级AI聊天 私有化部署 模型无关 数据安全 合规性 上下文记忆 供应商解耦 企业服务 B2B SaaS 人工智能平台
用户评论摘要:用户普遍认可其解决企业数据隐私痛点的价值,并询问具体技术实现(如数据驻留、与开源方案对比)、部署复杂度及市场定位。主要建议涉及官网页面结构需更清晰。开发者积极回复,解释了数据流转机制与产品优势。
AI 锐评

LyzrGPT切入的并非技术蓝海,而是一个明确且棘手的市场缝隙:企业日益增长的生成式AI需求与严苛的数据主权、合规要求之间的根本矛盾。其宣称的“私有化”和“模型无关”是同一枚硬币的两面,核心价值在于充当企业AI堆栈中的“安全代理层”与“流量调度器”。

产品真正的犀利之处,不在于其前端聊天体验——这已是红海,而在于其作为中间件对底层模型复杂性的封装和对数据生命周期的严格控制。它让企业IT部门获得了一种“受控的灵活性”:既能灵活调用多个外部顶尖模型API以规避供应商锁定并优化效果,又能确保原始数据文件永不外流,仅将必要的上下文信息以受控方式发送。这直接回应了金融、医疗、法律等受监管行业的核心关切。

然而,其挑战同样鲜明。首先,它面临上下夹击:上有微软、谷歌等云巨头提供绑定其模型的企业级安全方案,下有Open WebUI等开源项目搭配LLM代理的DIY方案。其竞争力必须体现在部署与集成的平滑度、长期上下文记忆的工程稳定性以及真正智能的模型路由算法上。其次,评论中关于部署复杂性和数据驻留的疑问,恰恰点中了这类产品的命门——在复杂异构的企业IT环境中,“内部部署”的实现成本和难度可能远超预期,这需要团队具备深厚的企业服务与集成经验。

总体而言,LyzrGPT若能在产品上扎实地兑现其安全承诺,在市场上精准定位那些对数据极度敏感、且具备一定技术能力但不愿从头自研的中大型企业,它便有望从一个功能型工具,演进为企业核心的、合规的AI能力基座。它的成功不取决于AI技术本身有多前沿,而取决于对企业治理规则的理解有多深刻。

查看原始信息
LyzrGPT
LyzrGPT is a private, enterprise-grade AI chat platform built for security-first teams. Deploy it inside your own ecosystem to keep data fully private. Switch between multiple AI models like OpenAI and Anthropic in the same conversation, avoid vendor lock-in, and retain secure contextual memory across sessions. Built for enterprises and regulated industries.
Hey Product Hunt 👋 We built LyzrGPT because enterprises told us the same thing again and again: “We want ChatGPT-level intelligence, but we can’t risk our data leaving our environment.” LyzrGPT is a private, enterprise AI chat platform that runs inside your own ecosystem. Your data stays with you. No vendor lock-in. Full control. You can also switch between multiple AI models (like OpenAI or Anthropic) within the same conversation and maintain secure, long-term context across sessions. We’d love your feedback: 1. What’s stopping your org from adopting AI chat today? 2. What security or compliance concerns do you face? Happy to answer questions in the comments. Thanks for checking us out 🙌
8
回复

@mohammed_faraaz_ahmed congrats on the launch! Are you targeting Europe? Europe could be a great market for this

0
回复

Found great use cases for enterprise and individual uses, do check it out!!

4
回复

So many Enterprises can benefit from this! I have seen a couple of them spending months on R&D and then finally creating a somewhat okay, brittle internal solution. This will not only reduce that time drastically but also the employees finally get to use something that actually works and do not have to end up using the public softwares to get work done. Kudos to the team for such a good job!

3
回复

@pradipta_ghoshal Thank you for the kind words! That’s exactly the gap we’re addressing.

2
回复
When a buyer compares you to rolling their own stack with an open-source UI + an LLM proxy and buying a secure enterprise chat from a big vendor, what are the 2–3 decisive differences that reliably make them choose LyzrGPT—and where do you intentionally not compete?
2
回复

@curiouskitty Great question!
-> In a DIY stack, you pay all of the vendors individual plans and then setup the whole architecture, would you do that or rather top up just one form of credits?
-> Not only that, LyzrGPT can automatically switch the model for the exact use-case, giving you the best output at all times.
-> With the import memory feature, no more filling in long context messages, once you import your previous memory from our memory pocket modal, no matter what model you choose, your memory is present and up to date!

1
回复

When you say data stays on your server, does that mean the files remain there but relevant information is still sent to GPT or Anthropic as context? How does that work?

2
回复

@hashif_habeeb 

The actual text content of relevant memories is sent to the LLM for inference - that's how AI works. But:

  1. We control what gets sent

  2. We control how much context

  3. Original files never leave our infrastructure

So essentially yes, your data lives with us, we just ask the AI questions about it.

0
回复
A great idea and really understands business needs for compliance. Checked a bit and the fact that you can create these agents or employees to handle tasks that usually are handled by humans without compromising on compliance or quality is amazing. I believe in the idea. Just as a feedback, I believe the landing page is quite dense and can benefit from a clearer structure. Wish you all the luck
1
回复

@zahran_dabbagh Thank you so much for taking the time to check us out and for the thoughtful feedback really appreciate it.Totally hear you on the landing page feedback as well. We’re actively iterating on the structure to make the value clearer and less dense, especially for first-time visitors. Inputs like this genuinely help us improve.

Hope you’ll continue following our journey

1
回复

@zahran_dabbagh Are there any other user experience shares

0
回复

Wow, LyzrGPT looks amazing! Love the model-agnostic approach. How does it handle data residency requirements across different cloud providers when switching between models? Super curious!

0
回复

Congrats!! it looks great! how does this compare with OpenWebUI?

0
回复

Running the AI chat inside the company’s own ecosystem is a big trust unlock.

Curious how difficult deployment typically is for enterprises with complex infra. Is this closer to plug-and-play or a guided rollout?

0
回复
#9
Cursor AI Mobile
Сontrol Cursor AI directly from your iPhone
130
一句话介绍:Cursor AI Mobile 是一款iOS伴侣应用,允许开发者在移动场景(如通勤、会议间隙)远程向Mac上的Cursor IDE发送指令,实现与AI编程助手的持续交互,解决了开发者离开主力工作机后工作流中断的痛点。
Developer Tools Artificial Intelligence Vibe coding
AI编程助手 移动伴侣应用 远程控制 开发工具 工作流协同 iOS应用 CLI集成 实时同步 效率工具
用户评论摘要:用户普遍认可其扩展工作场景的价值。主要反馈包括:赞赏其移动化能力;询问CLI处理大文件项目的性能;质疑其相较于远程桌面或通用聊天应用的独特必要性。
AI 锐评

Cursor AI Mobile 的本质,并非简单的远程控制或聊天转发,而是一次对“开发环境”定义的试探性重构。它试图将“编程”这一高密度、强上下文的认知活动,拆解为可异步、跨设备执行的“指令-结果”循环。其真正价值不在于技术实现的复杂度(通过CLI桥接),而在于它精准地切入了一个新兴的、模糊的“轻量级编程时刻”场景——那些在通勤路上突然产生的调试灵感、在会议间隙需要紧急查看的代码逻辑。它赌的是,开发者愿意为这种“思维连续性”和“即时捕捉”支付额外的操作成本与上下文切换代价。

然而,其面临的质疑也直指核心:在已有成熟远程桌面和SSH方案的情况下,它提供的价值增量是否足够厚?评论中的性能疑问和场景必要性拷问,正是其需要跨越的鸿沟。它可能不是通用解决方案,而是为深度依赖Cursor AI工作流、且对“碎片化时间利用”有极致要求的开发者量身定制的“效率杠杆”。它的成功与否,将取决于其能否在移动端提供足够精准、低延迟的上下文保持能力,以及能否培育出区别于“在手机上费力看代码”的真正移动端原生交互范式。否则,它极易沦为一个小众的、可有可无的玩具。

查看原始信息
Cursor AI Mobile
Access your AI coding assistant anywhere with Cursor Mobile - the companion app for Cursor IDE on iPhone and iPad. Send prompts from your device that run instantly on your Mac through the Cursor CLI, with results and chat history syncing back in real time for a continuous workflow.

Congrats on the launch! Love how Cursor Mobile extends the IDE to iOS so devs can review, debug, and iterate with their AI pair from anywhere, without breaking the desktop workflow.​

1
回复

Wow, Cursor AI Mobile is a game changer! Accessing my Cursor IDE on the go is HUGE. How well does the CLI handle projects with large file sizes when prompting from mobile? Super curious!

0
回复
What specific “away-from-your-laptop” moment (on-call, commuting, meetings, etc.) made you realize a mobile companion was necessary, and what evidence convinced you this wasn’t solved well by remote desktop/SSH or just using a generic chat app?
0
回复
#10
GitHub Space Shooter
Visualizes GitHub contribution graphs as a Space Shooter
129
一句话介绍:将枯燥的GitHub贡献图动态转化为复古太空射击游戏画面,在开发者展示个人编程活跃度或寻求社区趣味互动时,提供了一种极具娱乐性的可视化方案。
Developer Tools GitHub Data Visualization
GitHub工具 开发者趣味 数据可视化 游戏化 个人品牌 开源文化 极客娱乐 无实际用途 情怀项目 社交分享
用户评论摘要:用户普遍认为项目有趣、有创意,带来了欢乐。主要反馈是肯定其“纯粹为乐趣”的出发点,并欣赏其对经典游戏《太空入侵者》的情怀致敬。核心建议是希望推出可玩版本,而非仅静态图像生成。
AI 锐评

在效率至上的开发者工具红海中,“GitHub Space Shooter”是一款典型的“无用之用”反叛式作品。它刻意宣称“不解决任何问题”,恰恰精准击中了当前工具类产品过度强调功利价值的行业疲劳点。其真正价值并非功能层面,而是文化符号层面:首先,它充当了开发者身份的趣味勋章,将隐性的贡献记录转化为显性、可传播的极客梗图,满足了在社交媒体进行轻量化专业身份展示的需求。其次,它以“太空入侵者”这一文化符号为纽带,在怀旧与创新之间建立了情感连接,唤起了资深开发者的共鸣。

然而,其犀利之处也揭示了其天花板。产品目前停留在“图像生成”,与评论中期待的“可玩”存在体验断层,这暴露了其核心矛盾——是做一个有趣的“玩具”,还是真正具有交互性的“游戏化”工具?若停留在前者,其热度注定如流星般短暂,仅是一次巧妙的社交媒体营销;若迈向后者,则需深入思考如何将GitHub提交行为更深度地绑定游戏机制,例如将不同的提交类型(如修复bug、新增功能)对应不同的敌人或武器,从而将“可视化”升级为“可交互的驱动”,从单次娱乐消费转向长期习惯培养。当前版本更像一个聪明的概念验证(PoC),其成功反映了社区对轻松、人性化表达方式的渴望,但若要持续生存,必须回答“有趣之后,然后呢?”这一终极问题。

查看原始信息
GitHub Space Shooter
The project does not solve any problem. But just turn your github contribution graph into a space shooter battle field 🚀. You can either use the web app for a one time generation or use the github action for a schedule based generation.
The project does not solve any problem, and I built this purely for fun. Hope someone find this interesting!
4
回复

@zane__chen Hahaha, that’s so interesting. While everyone else is busy promoting their own products, what you said just made my day.

0
回复

@zane__chen Very cool, made me laugh :). How many engineers today even remember space invaders......

1
回复
@zolani_matebese Glad you laugh 😂. Thanks!
0
回复

I love it, playable version next?

1
回复

@janschutte Thanks! And Great Idea!

0
回复

Such a fun launch—turning GitHub contributions into a playable space shooter is a brilliant way to gamify dev activity.

1
回复
@zeiki_yu Thanks!
0
回复

I love it.

0
回复
0
回复
#11
Notto
AI dictation, meeting notes & invisible chat overlay
119
一句话介绍:Notto是一款桌面端隐形AI覆盖层应用,可在不切换应用的情况下,实现AI对话、语音转文字和会议记录,解决了用户在深度工作或会议中因切换应用而中断心流和注意力的痛点。
Productivity Meetings Artificial Intelligence
AI生产力工具 桌面覆盖层 语音转录 会议笔记 隐形助手 无干扰工作流 跨应用操作 实时AI聊天
用户评论摘要:用户赞赏其“隐形覆盖”理念能减少干扰,但新用户可能因UI过于隐蔽而感到困惑。存在快捷键冲突、安装包过大、崩溃等严重技术问题。开发者积极回应,承诺修复并优化。有用户质疑其与Claude Desktop等竞品的差异及长期生存能力。
AI 锐评

Notto的核心理念——“隐形覆盖层”——是一个极具洞察力的产品设计赌注。它精准地瞄准了现代知识工作者在多任务与深度专注之间的根本矛盾:对AI辅助的即时需求与切换上下文带来的高昂成本。其真正价值并非在于AI能力本身(这实为一种可被快速复制的通用能力),而在于试图重构人机交互的界面范式,将工具从需要被“打开”和“管理”的应用程序,转变为即需即用、用完即走的“环境层”。

然而,这一理念也带来了产品层面的双重挑战。其一为技术实现风险:评论中暴露的快捷键劫持、崩溃、资源占用过大等问题,并非普通bug,而是其作为底层覆盖层与操作系统及其他应用争夺控制权时必然遭遇的复杂性问题。这直接关系到产品的稳定性和可用性底线。其二为用户认知与习惯风险:“隐形”在带来流畅感的同时,也可能导致产品功能的“不可见”和“不可发现”,这与其作为工具需要被用户主动调用的本质相悖。评论中关于新用户困惑的反馈,正揭示了这一内在矛盾。

其长期生存能力确实面临双重挤压:横向有Claude Desktop等同类桌面AI工具的竞争,纵向则需警惕未来操作系统级AI助手(如升级后的Siri)的降维打击。Notto的护城河在于其对“深度工作流无缝集成”这一垂直场景的极致专注和体验打磨。若能在解决初期技术债后,持续优化触发逻辑、自定义程度与系统兼容性,它有望成为一类专注于“无干扰AI交互”的细分工具代表。反之,若无法在稳定性和体验上建立绝对优势,其“隐形”的优雅理念,恐将迅速被更稳定或更强大的平台型产品所淹没。

查看原始信息
Notto
The invisible AI overlay for your desktop. Transcribe meetings, dictate text and chat with your notes without switching apps. Works anywhere you do.
Hey Product Hunt! 👋 I’m Matei, the maker of Notto. We've all been there: You're in a deep work session or a meeting, and you need to ask AI a question, draft a quick email, or jot down a thought. But the moment you switch tabs to open ChatGPT or your notes app, you lose your flow. You break eye contact. The focus is gone. I built Notto to fix this. Notto is an invisible AI overlay that sits directly on top of your screen. It brings the power of AI right to your fingertips without cluttering your workspace. It allows you to: 💬 Chat with AI Invisibly: Ask questions, get answers, or brainstorm ideas while keeping your main window (or video call) in focus. 🎙️ Dictate with Speed: turning your voice into text for emails or documents instantly. 📝 Take Seamless Notes: Jot down action items or thoughts without ever looking away. The best part? It feels like a natural extension of your screen, not another clunky app to manage. I’d love for you to try it out and let me know what you think. Does the "invisible" overlay help you stay in the zone? Happy hunting! 🚀
2
回复

@anghelmatei Congratulations on the launch — wishing you great traction ahead!

If you need help with branding, UX/UI, MVP, or web design, Opedia Studio delivers scalable digital experiences for web and mobile apps.

0
回复

@anghelmatei Hi Matei, Congrats on the launch. How does this differ from cluade (and others) desktop? Am I not getting it or are you competing directly on being better?

1
回复

Huge congrats on the launch! Love the invisible overlay approach for frictionless AI note‑taking and dictation—feels like a natural fit for busy, multi‑app workflows on desktop.

2
回复

@zeiki_yu Appreciate the support, Zeiki! 🚀 'Frictionless' was definitely our main goal during development. We really wanted it to feel like a natural layer on the OS rather than just another app to manage. Hope it helps your workflow!

1
回复

Wow, Notto looks amazing! That invisible chat overlay is genius. How does it handle privacy when Im sharing my screen? Super curious!

1
回复

@jaydev13 Hey! So glad you like the overlay! Privacy was one of the biggest challenges we wanted to solve with Notto. To handle screen sharing, we built a 'Screen Capture Protection' toggle (found in the Stealth Mode settings). When it's on, Notto uses low-level OS APIs to tell the system to exclude those windows from any capture. The result is pretty magical: you can see the AI chat and your notes perfectly on your monitor, but for anyone watching your screen share or recording (via Zoom, Teams, OBS, etc.), the windows are completely invisible - they don't even show up as a black box! It’s currently available as part of our Ultra plan. 😉

1
回复

The “invisible overlay” ethos is a double-edged sword.

On one hand, it minimizes friction and clutter. On the other, it can sometimes feel too subtle — especially for new users who aren’t sure how to trigger it or where to look for controls. The simplicity that makes Notto elegant can also make it invisible in a way that feels confusing until you’re familiar with the UX.

1
回复

@kimberly_ross This is such a thoughtful take, Kimberly. You are 100% right.

Designing for "invisibility" is actually much harder than designing a standard interface because, as you said, users can sometimes forget the app is there or how to summon it!

We are trying to balance this with intuitive keyboard shortcuts and a very subtle "summoning" bar, but it is definitely a constant balancing act between "clean" and "confusing."

Thanks for the feedback!

1
回复

Sounds really interesting, so it's OS native, right? and is it compatible with all OS?

1
回复

@khashayar_mansourizadeh Glad you like the concept, Khashayar!

Yes, it is a desktop application (so it works over any other app, not just in the browser). We currently support macOS, Windows and Linux.

Since it creates a transparent layer over your screen, we had to go through the app route to get the performance right. Let me know if you give it a try!

2
回复
Congrats on the launch 👍 I just checked that and I love it. I believe that the app could benefit more from the idea that you put a bit subtly ”just asking that one question mid meeting”. This should be more dominant because this is a pain everyone have. But that’s my perspective only. However just upvoted 🚀
1
回复

@zahran_dabbagh Thanks for the support and the sharp feedback, Zahran! 🙌

You are spot on. That specific moment of anxiety—when you need to look up an answer mid-meeting but don't want to break eye contact or start typing furiously—was actually the spark that made me build Notto.

I’m going to take your advice and try to highlight that specific "mid-meeting helper" use case more prominent on the landing page. It's the feature I love most too!

Really appreciate the insight!

2
回复

I really wanted to like Notto, but within 5 minutes I ran into so many bugs it was unusable:

  • chat shortcut triggered a meeting usage. I was at 5 meetings transcribed when I had no meeting apps open within 2 minutes.

  • 783MB install size!

  • crashed after first run after granting all permissions

  • can't disable shortcuts – I have many other shortcuts assigned to the same combos

  • at one point it took over my spacebar, every hit opened the overlay

  • missing setting to change overlay transparency - I do not want to remember another shortcut to do this, and it's not a setting I'd want to change very often

  • overlay never worked

Running on MacOS 26.2 on an Nov 2024 M4

If it worked as described, I'd use it daily.
I also suspect this is one of the apps that will soon be replaced by Siri / Google Assistant when they get upgraded.

0
回复

@visle Hey Victor, first of all, thank you for the brutal honesty. This is exactly the kind of feedback that helps us move towards being a more reliable tool. I’m genuinely sorry the first experience was so friction-heavy.

I’m currently working on pushing a hotfix to address all of your points. Here is where we stand on your list:

  • Meeting Detection & Shortcuts: The 'spacebar takeover' and phantom meeting counts are bugs in our input-listening logic. We are fixing this so Notto stays strictly out of the way until summoned.

  • Customization: We’re adding the ability to disable shortcuts so they don't clash with your existing workflow.

  • Install Size: We are working on optimizing the install size to shave off ~250MB-350MB

  • Stability : I’m currently investigating the crash you experienced during the initial permission grant. It sounds like a conflict with the native API permissions on the newer M4 architecture, and I'm prioritizing a fix to ensure a smooth first-run experience.

A quick clarification to help me debug:

  1. Transparency: There is actually a slider in the Settings page, but your comment proves it’s not discoverable enough. I’ll be moving it to the top of the UI.

  2. 'Overlay never worked': Did the UI fail to appear entirely, or did you just not get an AI response back after prompting?

I’ve credited 1 month of Ultra to your account as a thank you for this breakdown. I'd love for you to give it another shot in a few days once the update is live.

As for Siri/Google replacing us - you might be right! Our goal in the meantime is to provide a much more specialized, 'invisible' workflow for power users that general assistants often overlook.

0
回复
#12
Manus Meeting Minutes
From in-person meeting to finished work in one flow
115
一句话介绍:一款专注于线下会议的AI工具,通过录音、说话人识别和智能摘要,将实体会议讨论一键转化为演示文稿、网站等交付成果,解决了会后信息整理与执行落地的效率痛点。
Productivity Meetings Artificial Intelligence
AI生产力工具 会议记录 语音转文字 说话人识别 智能摘要 内容生成 工作流自动化 线下会议 会后执行代理
用户评论摘要:用户肯定其聚焦线下会议及会后执行的独特定位,认为“会后代理”是竞争关键。主要问题集中于技术实现细节:说话人识别在多人同时讲话或嘈杂环境下的准确性,以及仅依靠手机麦克风的录音效果。
AI 锐评

Manus Meeting Minutes 的野心,不在于成为另一个“更好的录音笔”或“更准的转录工具”,而在于试图定义“会议”的终点。当前市场多数工具止步于“记录与回顾”,而Manus宣称要直达“产出与执行”,这击中了知识工作流中最顽固的断层:从共识到成果的“最后一公里”。

其真正价值并非简单的语音转文字,而在于构建了一个以会议录音为原始输入的“内容生成管道”。通过说话人识别赋予上下文,再通过AI代理将非结构化的对话流,按需编译成结构化的交付物(PPT、网站、文档)。这本质上是将会议本身“产品化”,把模糊的讨论瞬间封装为可传播、可迭代的资产。

然而,其宣称的“无缝工作流”面临双重考验。首先是技术可靠性:线下会议嘈杂的声学环境、多人插话的语音分离、手机麦克风的物理限制,都是其核心功能“说话人识别”的噩梦。若此基础不稳,后续所有高级生成都是空中楼阁。其次是市场定位的“窄”与“深”:专注线下会议虽避开了与Zoom、Teams等巨头的直接碰撞,但也主动放弃了规模更大的线上会议市场。这是一场高风险的押注,赌的是线下场景的痛点足够痛,且其解决方案的技术壁垒足够高。

评论中“忘记OpenAI”的说法略显夸张,但揭示了一个趋势:当AI能力趋于同质化,胜负手正从“模型能力”转向“工作流洞察”。Manus的挑战在于,它必须同时是顶尖的声学工程师、精准的NLP专家以及深刻的生产力流程重构者。若能跨过技术鸿沟,它或许真能开启“会后代理”的新品类;若不能,则可能只是一个功能华丽但使用场景脆弱的精致玩具。

查看原始信息
Manus Meeting Minutes
Introducing Meeting Minutes—record in-person meetings, get AI-powered summaries with speaker recognition, and turn discussions into presentations, websites, and deliverables in one seamless workflow.
Hi everyone! It is a fascinating move that Manus is focusing on in-person (IRL) meetings instead of the usual online ones this time. I resonate deeply with the insight here: A meeting is not the final stage, it is just the starting point of context. The end of a meeting is actually when the real work begins. Manus captures this perfectly. It takes that rich context and immediately bridges the gap to the next step, whether that is writing a follow-up email, generating a slide deck, or building a web page. It feels like the most logical evolution for their agentic platform. We have mastered the "summary," but the Post-meeting Agent—the one that actually executes the deliverables—is the Holy Grail of this wave of competition.
2
回复

@zaczuo Really like the framing here — meetings aren’t the finish line, they’re the start of context.
Turning an in-person recording into speaker-tagged notes and then directly into real deliverables (deck / doc / webpage) is the part most “meeting tools” never cross. If this reliably bridges that gap, it’s a big step toward a true post-meeting agent.
Congrats on the launch 👏

0
回复

Congrats on the launch! Love how Manus goes beyond chat to truly turn high-level intent into autonomous, multi-step execution across work and life.

1
回复

@zeiki_yu Congratulations on the launch! Wishing you strong momentum and continued success ahead.
If you ever need support with branding, UX/UI, MVP development, or web design, Opedia Studio creates scalable digital experiences for web and mobile applications.

0
回复

Wow, Manus is seriously cool! The Meeting Minutes feature is a game-changer. How well does the speaker recognition handle multiple people talking at once?

0
回复

Manus are unique. I literally forgot about OpenAI and now use them for research / agentic mode.

0
回复

Huge claim, but the Meeting Minutes bit is the hook. My in-person notes are chaos and I forget who said what. If it nails speaker tags and turns it into a draft deck/site, that’s useful. Curious how it handles noisy rooms and just a phone mic.

0
回复
#13
StackPatch
Patch authentication into your Next.js app with one command
112
一句话介绍:StackPatch是一款CLI工具,通过一条命令为Next.js项目注入生产就绪的认证代码,将开发者从重复、繁琐的认证配置中解放出来。
Productivity Open Source Developer Tools GitHub
开发者工具 Next.js生态 身份认证 命令行工具 开源 本地化部署 代码注入 生产力提升
用户评论摘要:用户普遍赞赏其“无SaaS、无锁定、代码完全自有”的理念及优秀的开发者体验。主要问题与建议集中在:如何处理复杂项目集成边界;未来是否支持RBAC、多租户;补丁模式能否泛化至其他框架或领域;以及更新维护机制。
AI 锐评

StackPatch的锋芒在于其精准的定位与克制的哲学。它没有选择构建又一个全托管的认证SaaS,而是反其道行之,将“代码所有权”和“无魔法”作为核心卖点,直击资深开发者在重复劳动与供应商锁定之间的双重痛点。其本质是一个高度特化的代码生成器,但通过“补丁”这一概念和CLI的极简交互,巧妙地将可控性和便捷性结合在一起。

然而,其深层价值与潜在风险同源。将标准化认证逻辑以真实文件形式注入项目,固然赋予了开发者完全的掌控权和透明度,但也将后续的更新、安全维护责任完全转移给了开发者。这要求团队具备相应的运维能力,否则“生产就绪”的初始状态可能随着时间推移而腐化。创始人“安全默认 > 广泛覆盖”的支持原则是明智的,这避免了工具在追求泛用性中变得臃肿和不可预测,但也必然限制了其在复杂、非标项目中的适用性。

从生态位看,它介于NextAuth.js等配置型库与Auth0等云服务之间,提供了一个折中选项。但其长期生命力将严重依赖于对Next.js框架演进的紧密跟进,以及“可选补丁”(如RBAC)生态的构建。它更像一个精心打磨的专用手术刀,而非瑞士军刀,在追求效率和控制的特定场景下表现出色,但并非所有团队的普适解药。

查看原始信息
StackPatch
StackPatch is a CLI that patches production-ready authentication into Next.js projects. Run one command and get Google & GitHub auth, protected routes, and session handling all as real files inside your repo. -No SaaS -No lock-in -Fully open source.
Hey Product Hunt 👋 What inspired me to build StackPatch was pure repetition. Every time I started a new Next.js project, authentication wasn’t hard, it was just the same setup over and over again: NextAuth config, routes, providers, protected pages, environment variables. I didn’t want another hosted service or abstraction. I wanted a way to drop production-ready auth directly into my codebase and fully own it. That’s how StackPatch started. My approach evolved from trying to “simplify auth” to something more honest: patching real, editable files into the project using a CLI. No magic, no runtime dependency just code you can read, modify, or delete. For this launch, I focused on keeping the scope tight: one command, clear output, and a demo that shows exactly what gets added. I’d love feedback from the community: • Does this solve a real pain for you? • What would you want to patch next? Happy to answer any questions 🙂
3
回复

@darshitdev Congratulations on the launch — wishing you great traction ahead!

If you need help with branding, UX/UI, MVP, or web design, Opedia Studio delivers scalable digital experiences for web and mobile apps.

Logo & Brand Guidelines: $599

Landing Page Design: $399

0
回复

@darshitdev This is such a clean idea — auth isn’t “hard”, it’s just painfully repetitive.
I love that StackPatch ships real, editable files into the repo (no SaaS, no lock-in, no magic runtime), so teams fully own the code from day one. That’s the kind of DX that actually fits production work.
Congrats on the launch — this feels like it’ll become a default starting point for a lot of Next.js projects 👏

1
回复

Congrats on the launch! Love how StackPatch patches battle-tested auth into existing Next.js apps with one command—great DX, zero lock-in, and perfect for production-focused teams.

2
回复

@zeiki_yu Thanks more frameworks and feature coming soon in this open source

0
回复

@zeiki_yu congrats Zeiki. How do you handle updates and maintenance?

1
回复
Next.js projects vary a lot (App Router vs older patterns, monorepos, path aliases, existing providers/middleware). What’s the hardest integration edge case you’ve had to handle so far, and how do you decide when to support a pattern vs. intentionally say “not supported”?
1
回复

@curiouskitty Great question!! The hardest cases so far have been projects that already have partial auth setups (custom middleware, existing providers or mixed App Router + legacy patterns). Those are tricky because you’re not starting from a clean slate.

My rule is: StackPatch should never guess intent.

If a pattern can be:

-reliably detected

-patched without overwriting custom logic

-and leave the project in a predictable state

then I support it.

If it requires assumptions (for example: custom auth flows, deeply coupled middleware, or heavily modified folder structures), I intentionally fail fast and mark it as “not supported” rather than silently doing something risky.

The philosophy is: safe defaults > broad coverage. I’d rather support fewer patterns well than many patterns unreliably.

Over time, I’m expanding support based on real-world project structures I see and test against.

0
回复

Looks interesting, I feel like there are a couple good use cases for this. I'm wondering how flexible this is, do you have generic way of defining "patches". Could it also work for other domains and apps that are not Next.js? Is it / could it be combined with AI to automatically patch projects?

1
回复

@wilco_kruijer1 yes currently i am handling use cases for the existing app/project without messing your file structure introducing good design principle and you can revert the changes if something is not up to the mark all the files will be replaced as it is as well as the new project for other domains i am working on it will be out soon! I am not combining this with AI as AI needs lot of iteration to make this work and also AI is not updated with the latest doc changes and it can change file randomly or create new files so lets see

0
回复

This is awesome — I love that it just patches real files into the codebase.

No magic, no hidden stuff, just code you actually own.

Quick question: do you plan to support Role-Based Access Control in the future, or is the focus staying purely on authentication for now?

Can’t wait to try this on my next Next.js project

1
回复

@mrpop Thanks a lot really appreciate that 🙂

The core focus is staying authentication-first and composable.

I don’t want StackPatch to turn into a heavy auth framework.

RBAC is something I’m thinking about as an optional patch rather than a default something you add only if you need it and fully own the code.

Would love to hear how you usually handle roles in your projects.

0
回复

Wow, StackPatch looks amazing! Love that it adds auth as real files in my repo – huge for customization. How does it handle multi-tenancy scenarios, say, separating user data by organization?

0
回复
#14
AI dication by Snaply
Free, 100% local, AI dictation for Mac
101
一句话介绍:一款完全免费、100%本地运行的Mac端AI语音听写工具,通过离线处理解决了用户对隐私安全和高昂订阅费用的核心痛点,适用于需要快速、私密地进行文字输入的各种场景。
Productivity Artificial Intelligence Apple
AI语音输入 本地离线处理 隐私安全 macOS工具 免费软件 语音转文字 效率工具 听写软件
用户评论摘要:用户普遍赞赏其本地化带来的隐私优势。主要反馈集中在功能需求上:1. 请求增加自定义词典(如防止特定词误转写)和快捷键自定义;2. 询问是否支持接入用户自建的本地语音模型端点;3. 期待Windows版本;4. 询问多语言支持。开发者互动积极,对高级功能持开放探索态度。
AI 锐评

Snaply精准地切入了当前AI应用市场的两个敏感神经:隐私与成本。它并非简单提供一个免费替代品,而是通过“100%本地运行”这一技术选择,将“免费”从营销噱头升维为核心的产品价值观和信任基石。这直接击中了那些对云端数据处理心存疑虑、又厌倦了SaaS模式持续订阅压力的专业用户。

然而,其真正的挑战在于“本地”这把双刃剑。本地化固然带来了隐私和速度的优势,但也意味着性能完全依赖于用户本地硬件(如Apple Silicon的普及度),模型更新和功能迭代将受限于本地算力,难以像云端服务那样快速部署更强大的模型。从评论中高级用户对“接入自建模型”的强烈需求可以看出,当前内置的Whisper模型可能无法满足所有专业场景的精度和定制化要求。

产品的未来走向呈现一个关键分岔路:是坚持极简、易用的“开箱即用”工具定位,服务于大众用户;还是向技术爱好者与专业用户敞开怀抱,提供模型选择、API接入等高级功能,成为一个可扩展的本地AI输入平台?开发者的回复显示出在这两者间权衡的迹象。此外,将Electron应用与大量原生Swift代码深度绑定的技术路径,虽然优化了Mac体验,却也为其跨平台扩张(如回应中明确暂无Windows计划)筑起了高墙,这可能使其在赢得特定群体喝彩的同时,牺牲了更广阔的市场可能性。它的成功,将取决于能否在“本地隐私堡垒”的狭窄赛道上,构建出足够深且不可替代的体验护城河。

查看原始信息
AI dication by Snaply
Stop typing and just speak. Snaply brings instant AI dictation to every app and website on your Mac. Because it runs 100% locally on-device, your data remains completely private.🔒 Best of all? It’s entirely free for individuals.💰
Hey Product Hunt! 👋 I’m Giacomo, the developer behind Snaply. I’m incredibly excited to share what I’ve been working on. To be honest, the story of Snaply starts with me being a user of other dictation apps. I really loved the concept of AI voice typing, but I hit two major roadblocks that I just couldn't get past: The Cost: Existing tools were charging up to $12/month for individuals and $40/month for enterprise. That felt like a lot for a utility. The Privacy: I wasn't comfortable with the idea that my voice recordings and private thoughts were being processed on someone else's servers and potentially used for data collection. I wanted a tool that respected my privacy and my wallet. So, I built Snaply. 🚀 What is Snaply? Snaply is an ultra-fast AI dictation tool for macOS that runs 100% locally on your device. Because we don't need to pay for expensive cloud processing servers, I am making Snaply 100% Free for individuals. ✨ Why you’ll love it: 🔒 100% Private & Offline: Your voice data never leaves your Mac. It runs on local Whisper models, so you can dictate sensitive emails or confidential code without worry. ⚡ Incredibly Fast: Since there is no latency sending data to the cloud and back, Snaply is blazing fast and responsive. 💸 Free Forever (Individuals): No monthly subscriptions for personal use. Just download and go. 🧠 Smart Grammar: It doesn't just transcribe; it polishes. It fixes grammar and punctuation on the fly so you don't have to edit. How to use it (it’s super simple): Download Snaply for Mac. Press your shortcut key (like Function or Command). Start speaking! Snaply types perfectly into Notion, Slack, VS Code, or wherever you are. I’d love for you to give it a spin and let me know what you think. Since this is the first release, your feedback means the world to me and will help shape the future of the app! Thanks so much! 🙌 Giacomo
3
回复
@venier Will check this out as I’m still interested in a local approach. Do you have the dictionary feature? So if I say what might get interpreted as “psalm” it would always be “SOM” instead. It would be hard to leave WisprFlow without that. Also customizing shortcut keys for quick and long mode.
2
回复
Congrats @venier! (the best name btw) 😉 I run a local server with an endpoint that exposes a few stt models. Since I see you’re planning to let the user pick external models, will I be able to set up my local endpoints as well?
2
回复

@igecky Haha, great name!

At the moment, we are planning to support external LLMs.

We have not yet considered external TTS models though. Which models are you hosting, and how does the API work?

Our goal is to keep the product very user friendly and oriented toward non technical users. That said, we could explore adding more advanced options that are somewhat hidden or optional for power users 🙌

1
回复
@venier I expose my local models via an openAI compatible api so it should be easy for you. Basically I behaves exactly as if it were Whisper endpoint
1
回复
@venier yeah I thought you were using parakeet under the hood because of the speed but since I have it already on my server, I was wondering if I could avoid to have to download it again on the client. I like to use different models and hopping from parakeet to whisper, depending on the use case, so it would be great to be able to integrate my local endpoints 
1
回复
This is exactly what I need, but I'm using Windows. When will a Windows version be available?
1
回复

Hey@riley_evan ! Thanks, I am glad you liked it.

Unfortunately, building a Windows version is not currently on the roadmap.

The app is built using Electron app, which by default would be multi-platform, but we added a lot custom Swift code to make the app run smoothly on Mac.

0
回复

Wow, this is amazing! Local AI dictation is a game-changer for privacy. Im curious, does Snaply support multiple languages besides English out of the box?

0
回复
#15
Hal9
Launch AI-Powered Products in 30 Days
94
一句话介绍:Hal9为不具备技术背景的创业者提供“专家指导+AI代理”的协作模式,在30天内将AI创意转化为可上线、安全且可扩展的MVP产品,解决了他们因技术资源匮乏、开发成本高昂或陷入无效“氛围编码”而无法快速验证市场需求的根本痛点。
Artificial Intelligence GitHub Consulting Vibe coding
AI应用开发平台 非技术创始人 MVP快速构建 专家指导服务 AI代理开发 30天产品上线 定制化AI产品 创业加速 技术外包替代方案
用户评论摘要:用户反馈积极,认可其价值主张。主要问题集中于:能否接手修复现有“氛围编码”项目、如何平衡深度定制与30天时限、如何处理与遗留系统集成、如何进行产品验证,以及创始人最常见的瓶颈阶段。创始人回复坦诚,指出瓶颈普遍存在,且多数人对自身需求判断有误。
AI 锐评

Hal9的叙事精准地刺中了当前AI创业泡沫下的一个真实痛点:无数怀揣想法的非技术创始人,在ChatGPT等工具的鼓舞下投身“氛围编码”,最终在技术可行性、系统架构、安全与规模化面前撞上南墙。其宣称的“专家指导AI代理”模式,本质上是将传统软件外包进行AI时代的概念重塑与流程再造。

产品的真正价值不在于其技术栈有多前沿,而在于它试图在“全自动AI开发”的幻想与“昂贵人力团队”的现实之间,找到一个务实的中间路径。它售卖的是“确定性”——通过专家介入来定义边界、控制风险,再以AI代理提升执行效率,最终打包成一个“30天交付”的确定性承诺和可预测的定价。这比单纯提供开发工具或纯人力服务,更能缓解非技术创始人的焦虑。

然而,其模式面临深层拷问:首先,“30天”的强约束是否会导致产品妥协为另一个标准化模板的变体,与“深度定制”形成内在矛盾?其次,其核心壁垒是AI工具链还是其专家团队?若是后者,其服务规模化和质量稳定性将面临挑战。评论中关于集成与验证的提问,也触及了其服务范围的边界——它可能擅长从0到1的绿地项目,但对改造现有复杂系统或替代真正的市场验证,能力恐怕有限。

总体而言,Hal9是AI应用开发民主化浪潮中的一个理性修正案。它承认了当前阶段纯AI协作的不足,并试图将人的专业判断重新置于关键节点。其成功与否,不取决于AI代理有多智能,而取决于其能否系统化地复制专家经验,将“指导”过程本身产品化、规模化,从而在效率与质量间找到可持续的平衡点。

查看原始信息
Hal9
Hal9 helps non-technical founders launch AI-powered products in 30 days without expensive teams or endless AI prompts. Instead of vibe-coding alone with robots, talk to our AI experts who scope your vision, refine ideas, guide AI agents to build secure scalable MVPs on our platform. It's the most competitive way to ship AI MVPs: experts guide (don't code from scratch), AI handles the heavy lifting, predictable pricing, no lock-in, full customization. Focus on growth while we handle the tech.
Hi ProductHunt! Javier here, founder of Hal9 👋 We help non-technical founders like many of you turn AI ideas into launched products in just 30 days without the usual dev headaches or crazy costs. No more endless prompt wrestling or solo vibe-coding dead ends. Our experts guide AI agents to build secure, scalable MVPs while you focus on users, growth, and fundraising. We’ve shipped dozens of real AI apps (chatbots, agents, apis, mobile apps, extensions) this way. Super excited to see what you all think. AMA!
2
回复

@javierluraschi congrats on the launch! Do you also help with fixing vibe coded apps or more starting from scratch?

1
回复

@javierluraschi Congrats on the launch Javier. So this is like aicofounder crossed with base44? How do you deal with product validation?

1
回复

Congrats on the launch 🙌🏻 Super interesting approach. Curious to know: at what point do founders usually hit the biggest blocker, ideation, scoping, or actually shipping the MVP with AI agents?

1
回复

@thomaxx0 very smart question! It's really all of them. We've helped them on each stage of the process. HOWEVER, if you ask founders, no one thinks they need help on ideation or scope, everything thinks that they just need a little bit of help to "finish" the MVP. Unfortunately, that's usually the tip of the iceberg. There are cases where we do help them finish the MVP, but in most cases, we work with them to understand why the MVP is not getting finished. Could be multiple issues: Scope creep, tech not viable, incorrect tech, right problem but wrong solution, solution too complex, and on and on. So while many would think they just need to "finish" their MVP, once they work with us they usually find ways to do it all over in a much smarter and targeted way that starts getting them customers faster. You want something simple that can get you your first customers, nothing else matters.

We have an article about this but we learn more about this every day: https://hal9.com/articles/why-ai-mvps-fail

0
回复
Hey Javier, that line about solo vibe-coding dead ends hits. Was there a founder you worked with who had a solid AI idea but got completely stuck trying to build it themselves?
1
回复

@vouchy ALL OF THEM! We have a few testimonials in our website hal9.com. For the most part, everyone hits a wall at some point and need help getting unstuck. Different founders hit a wall at different stages, some very early on and it feels like they are missing 80% of their vibe-coded product, others come at 90% and find out they can't get the tech or AI to perform closer to 100%.

0
回复

Hal9 looks amazing! Im really digging the promise of rapid AI product development. How well does it handle integration with existing legacy systems? Super curious!

0
回复

Very compelling value prop. How do you balance deep customization with the 30-day launch timeline, especially for products with more complex workflows or data requirements?

0
回复

Congrats on the launch! Love how Hal9 blends expert guidance with AI agents to ship real, secure AI MVPs for non-technical founders in weeks, not months.​

0
回复

@zeiki_yu Thank you! Yes we think fully embracing AI vibe-coding + AI expert is extremely important. At least today, working with an AI + Expert, is still way superior than just working with an AI. Maybe that will change in a few years, but not any time soon.

0
回复
#16
Flowdy
Turn product descriptions into shoppable links automatically
91
一句话介绍:Flowdy是一款AI驱动的工具,能自动将博客和产品描述中的关键词转化为可购物链接,在用户阅读内容时无缝弹出商品推荐,旨在解决内容流量转化率低的痛点,将读者直接转变为买家。
Marketing Artificial Intelligence E-Commerce
内容电商 SaaS工具 AI购物助手 Shopify插件 流量变现 无摩擦购物 个性化推荐 关键词变现 博客商业化 电商增效
用户评论摘要:用户反馈积极,认可其解决“内容有意向但无转化”的核心痛点。创始人详细解释了其“AI辅助而非自动驾驶”的差异化逻辑,强调商户对关键词和商品映射的强控制权,尤其适合利基市场。用户询问了在细分品类的适配性,并肯定了“流失收入可视化”概念的价值。
AI 锐评

Flowdy切入了一个精准且古老的电商痛点:内容与交易的割裂。其宣称的“零摩擦”和“神经流”个性化,本质是将传统的上下文广告与基于会话的即时推荐相结合,技术栈上并无颠覆性突破。其真正的锋利之处在于两点:一是产品形态上,以侧边抽屉而非跳转页面的方式实现“无感购物”,在体验上确实降低了转化门槛;二是价值主张上,直指商户最敏感的“流失收入可视化”,将抽象的流量损失转化为具体的美元数字,极大地增强了销售驱动力。

然而,其面临的挑战同样清晰。首先,它高度依赖商户自身的内容质量和关键词标记的精细度,在“AI建议”与“人工控制”之间,可能为商户增加了新的运营负担。其次,所谓的个性化基于单次浏览会话,在数据厚度和推荐精准度上,能否超越成熟的、基于全站用户数据的推荐引擎,需要打一个问号。最后,其商业模式本质是帮助商户“榨取”现有流量的剩余价值,而非带来新流量,市场天花板可能受限。

总体而言,Flowdy是一款优秀的“增效工具”,而非“革命性产品”。它在电商变现链条的最后一公里做了巧妙的微创新,用更优雅的方式实现了“即看即买”。但其长期价值不在于AI有多智能,而在于能否深度融入内容创作与商品运营的工作流,成为不可或缺的“提效插件”。在独立站生态中,它具备明确的生存空间,但若想突破边界,仍需思考如何从“流量榨取者”转变为“价值创造者”。

查看原始信息
Flowdy
Turn content into shoppable links. Flowdy uses AI to highlight keywords in your blogs and descriptions automatically. Most readers leave without buying. Flowdy fixes this. When clicked, instant product recommendations appear—boosting conversion without page reloads. Why Flowdy? ⚡ Zero Friction: Shop seamlessly while reading 🧠 Neural Flow: Personalized suggestions based on browsing history 📈 Monetize Content: Track revenue per keyword Turn readers into buyers today.

Most blog readers leave without buying. I built Flowdy to change this.

The problem:

Merchants write great content, but readers bounce when they're done. That's wasted traffic.

The fix:

Flowdy auto-highlights keywords in your text. Click → side drawer opens → products appear. No page reload. No friction.

🧠 What makes it different:

"Neural Flow" It remembers what you click. Tap "cotton" → next suggestions prioritize cotton products. Like a shop assistant that learns in real time.

My favorite feature:

Opportunity Radar It scans your store and shows: "This keyword gets 500 clicks/month but has no products linked. You're losing $2,000."

Built solo. Privacy-first. Works with any Shopify theme.

🙏 Question:

Does the "lost revenue visualization" concept resonate with you? Hanging out in comments—ask me anything!

1
回复

Wow, Flowdy looks amazing! Love the idea of zero friction shopping right in the blog. How well does the personalized suggestion engine adapt to very niche product categories?

1
回复

@jaydev13 

Thank you! Great question! Flowdy is actually designed to excel with niche

categories. Unlike AI-driven tools that guess what to show, Flowdy lets

merchants control exactly which products appear through keyword-tag mapping.

The merchant knows their niche best—so they set keywords like "organic cotton"

or "handcrafted leather," link them to product tags, and only those productsappear.

AI assists by suggesting keywords and optimizing display order, butthe merchant stays in control.

This "AI as assistant, not autopilot" approach means niche stores get precise,relevant recommendations without generic AI guesses.

0
回复

Yeah Kosuke! Making it easier than ever cause I see that problem. Blogs aren't made to convert and it's a big mistake. Luckily you're helping a lot with Flowdy and I'm sure many founders gonna see the same. All the best!

1
回复

@german_merlo1 

Thank you! I'm really happy to hear that.

​I felt that pain deeply too. I also realized that most recommendation engines focus too much on past history rather than the now.

I wanted to capture the immediate "heat" (intent) generated by the blog content and descriptions.

​Your comment is incredibly encouraging! Thanks again!

0
回复

Kudos on your launch. As a founder, this resonates. Content often drives intent but not revenue.
Making lost opportunities visible feels like the right thing.

1
回复

@anupamsingh0211 

Thank you so much!

Hearing that from a founder means a lot to me as a developer. You nailed it with "intent but not revenue"—bridging that exact gap was my goal. I'm just happy if Flowdy can help turn those lost opportunities into actual growth for merchants.

0
回复
#17
WizzyPoke
Instant vibration alerts across all browser tabs.
86
一句话介绍:WizzyPoke是一款浏览器扩展,通过发送跨标签页的屏幕震动与声音提醒,在异步在线沟通场景中,以轻量、有趣的方式解决“对方是否看到消息”的确认痛点,避免重复追问或无效通知。
Chrome Extensions Productivity Messaging
浏览器扩展 实时提醒 跨标签页通知 轻量沟通 团队协作工具 趣味互动 无账户登录 免费增值模式 产品猎手发布
用户评论摘要:用户正面评价其将消息确认转化为轻量有趣的互动,避免了通知骚扰。创始人透露灵感源于MSN“震动”功能,旨在实现有意图的提醒而非制造噪音。另有评论为创始人的首次发布与坚持所鼓舞。
AI 锐评

WizzyPoke的本质,是在过度通知的疲劳时代,一次对“有效触达”的复古式解构。它聪明地避开了争夺通知栏红点的军备竞赛,转而利用浏览器标签页这一相对空白的物理感知层(视觉震动+声音),将提醒行为游戏化、轻量化。其核心价值并非技术突破,而在于对社交压力的精准拿捏:用一个无需回复、略带戏谑的“戳一下”,替代了“在吗?”或重复发送消息的焦虑感,降低了沟通的心理成本。

然而,其产品形态也暴露出明显的天花板与风险。首先,其应用场景严重依赖于双方同时在线且均安装扩展的“配对”模式,这构成了极高的使用门槛,限制了其从趣味工具向通用协议的演进可能。其次,“震动”功能的新鲜感衰减后,可能从有趣的提醒沦为新的干扰源,如何定义“重要”与“滥用”的边界,产品本身并未提供机制化解决方案。最后,作为浏览器扩展,其生存受制于平台政策,功能稳定性与长期存续性存在隐忧。

总体而言,WizzyPoke是一个精巧的“痛点止痛片”而非“根治方案”。它证明了市场对低压力异步沟通存在需求,但其作为独立产品的长期价值有限。它更可能的出路,是作为一项功能被整合进Slack、Discord等成熟协作平台,或是成为特定垂类社区(如远程团队、游戏开黑群体)的内部趣味插件。创始人的“情怀”与用户的“怀旧”能赢得初次掌声,但要在红海中存活,需要更快地构建场景壁垒或探索更深层的协作价值。

查看原始信息
WizzyPoke
WizzyPoke is a browser extension that lets you send subtle screen shake to get someone’s attention when they miss your message. After pairing with a simple 6-digit code, you can send a nudge that triggers a vibration style screen shake with sound, even if they’re on another tab or website. Works across all tabs from messaging apps to video sites without accounts or logins. The free version includes limited daily nudges, with an optional premium upgrade for unlimited use.

Congrats on the launch! Love how WizzyPoke makes “did you see my message?” a lightweight, fun nudge instead of another notification to ignore.​

2
回复

@zeiki_yu  thanks a lot!

0
回复

Hey everyone, WizzyPoke inspired by the old MSN “nudge” feature a simple way to get someone’s attention when messages easy to miss. I wanted to bring that idea back in a more modern and controlled way: a lightweight browser extension that sends vibration style screen nudges across all tabs, once paired with a simple code. The goal isn’t to be noisy, but intentional helping people notice something important without adding more clutter. I’d love to hear your thoughts, use cases, or ideas for improving it.

1
回复

This is my first product and my first Product Hunt launch. I started vibe coding just 2 months ago with no prior experience at all. I genuinely loved the idea while building it, but when no one was using it early on, I honestly thought the product might just be bad and almost dropped it. Seeing the support and upvotes here really means more than I can explain. Thank you to everyone who took the time to check it out. It gave me real motivation to keep building and improving new products.

0
回复
#18
AI Mode API
Query Google’s AI Mode as a simple, programmable API
73
一句话介绍:AI Mode API 是一款浏览器扩展,它将谷歌AI模式转化为可编程的API端点,解决了开发者和研究人员需要自动化、结构化获取谷歌AI生成答案及来源信息的痛点。
Marketing Developer Tools GitHub
API工具 浏览器扩展 谷歌AI集成 数据抓取替代方案 自动化脚本 结构化数据 研究工具 开发者工具 开源项目 免费增值
用户评论摘要:开发者介绍了产品原理与免费、计划开源的状态。用户评论赞赏其结构化JSON输出,并提出了关于CAPTCHA处理机制的实际技术问题,显示用户关注其自动化流程的鲁棒性。
AI 锐评

AI Mode API 本质上是一个精巧的“协议转换器”与“合法化中间层”。其真正价值不在于技术上的高深,而在于敏锐地捕捉并填补了一个官方服务与自动化需求之间的灰色地带。谷歌AI模式本身未提供API,手动查询效率低下且难以集成。该产品通过浏览器扩展这一载体,将用户本地的浏览器会话转化为一个合规的请求代理,既绕过了官方API缺失的障碍,又避免了直接爬取可能面临的法律与反爬风险。

然而,其“通过你的浏览器”这一核心运行机制,既是亮点也是最大的阿喀琉斯之踵。它注定受限于单浏览器实例的性能与稳定性,难以承载高并发请求,所谓“用于研究和监控,而非高产量生产”的声明,实为对自身根本局限性的坦诚。CAPTCHA问题更是悬在头顶的达摩克利斯之剑,一旦谷歌加强验证,整个自动化流程可能瞬间崩溃。

长远来看,该产品的命运与谷歌官方的态度紧密绑定。它是一个在官方生态缝隙中求生的临时解决方案,其生命力取决于官方API的推出速度以及对其模拟行为的容忍度。开源服务器代码的承诺是明智之举,旨在将工具价值转化为社区信任,吸引早期采用者。在当前阶段,它是技术极客和特定研究场景下一款锋利但可能易折的“手术刀”,而非可规模部署的“工业引擎”。其成功与否,将是对市场时机、需求刚性以及开发者应对政策风险能力的综合考验。

查看原始信息
AI Mode API
What it does: Turns Google's AI Mode into a programmable API endpoint that runs through your browser. Install the extension, get an API URL, and start querying AI Mode from anywhere—your terminal, scripts, automation tools, whatever. What you get back: Every response includes the AI-generated answer, follow-up questions, and source links. All structured as JSON. No scraping setup, no complex tooling.
Hey — I'm Patt, I built this. I needed to query Google's AI Mode programmatically. No API existed. So I built a browser extension that gives you a private API endpoint tied to your browser. How it works: Install the extension → it connects to a relay server → you get a unique, private API endpoint. Send POST requests to it from your scripts, terminal, whatever. Your browser receives the query through the extension, the extension talks to AI Mode (no login required), and you get back JSON with the answer, follow-up questions, and source links. Important: Your endpoint is private — only you can query it. The extension only interacts with AI Mode, nothing else in your browser. The video: First 2 minutes is a live demo. After that, I walk through the technical details and what you need to know to use this seriously. Current status: It's running, it's free, I'm covering infrastructure costs. Planning to open source the server code so you can self-host. Rate limits apply since this runs through your browser session — built for research and monitoring, not high-volume production. I'm around for questions. — Patt
0
回复

Wow, this AI Mode API is awesome! The structured JSON response including follow-up questions is a game changer. Does the extension handle CAPTCHAs automatically or is that something wed need to manage in our scripts?

0
回复
#19
GROOVY
Universal Search and Signaling across LLMs
71
一句话介绍:GROOVY是一个面向AI智能体的统一仪表板,通过提供跨大语言模型的全局搜索与信号传递功能,旨在解决知识工作者在使用多个AI工具时因等待响应和频繁切换上下文而难以进入“心流状态”的痛点。
Productivity Artificial Intelligence
AI工作流整合 跨模型搜索 智能体协作 统一仪表板 生产力工具 心流状态 LLM管理 知识工作 上下文切换 AI代理
用户评论摘要:用户普遍认可产品解决“标签页混乱”和“上下文切换”痛点的理念,认为统一搜索与协调层是正确方向。主要疑问集中于其如何处理不同模型的提示词格式差异,技术实现上团队透露使用了Vercel AI SDK。
AI 锐评

GROOVY切入了一个真实且日益凸显的痛点:AI工具碎片化对“心流”的破坏。其价值不在于替代某个LLM,而在于试图成为AI工作流的“操作系统中间层”。产品提出的“跨LLM信号传递”概念颇具野心,这暗示了其未来可能的方向——不仅是统一界面和搜索,更是构建智能体间的通信协议与协作框架。

然而,其当前形态面临严峻考验。首先,“统一搜索”的技术壁垒极高,不同LLM的底层逻辑、能力边界和输出风格迥异,简单的API聚合难以带来真正连贯的体验,反而可能让用户陷入另一层“提示词工程”的泥潭。其次,产品定位略显模糊:是面向个人极客的效率工具,还是瞄准团队协作的基础设施?评论中提及的“新Slack for AI”的愿景,需要极强的网络效应和生态构建能力,这远非一个仪表板所能承载。

真正的挑战在于,随着各大模型厂商和Notion等生产力平台纷纷内置多模型切换与工作流功能,一个独立中间层的生存空间可能被迅速挤压。GROOVY若想突围,必须将“信号传递”做实做深,从简单的历史记录与调用转向定义智能体交互的元协议,成为复杂AI工作流不可或缺的“调度中枢”。否则,它可能只是一个过渡时期的优雅补丁,而非下一代AI工作的基石。

查看原始信息
GROOVY
"We are outperforming ourselves thanks to AI by at least 1.5x in anything knowledge work related now. Coding, sales, marketing, growth, funding, law, taxes. You can get very competent very fast. The main problem with AI tools right now is that most of them don’t let you reach flow state. You have to wait after prompting and you’re constantly context switching." - @yasser_elsaid_ Groovy is a unified dashboard for all your AI agents. Offering you universal search and signaling across LLMs.

Congrats on the launch! Groovy’s unified dashboard and cross-LLM signaling feel like a powerful way to keep agent workflows in flow state instead of tab chaos.

2
回复

@zeiki_yu Congratulations on the launch — wishing you great traction ahead!

If you need help with branding, UX/UI, MVP, or web design, Opedia Studio delivers scalable digital experiences for web and mobile apps.

0
回复

Wow, GROOVY looks amazing! The idea of unified search across LLMs is a game changer. How does it handle nuances in prompt formatting across different models? Super curious!

1
回复

@jaydev13 We use the super useful and amazing Vercel AI SDK

0
回复
It is clear that AI is making all of us more productive, but context switching isn't helping. There needs to be a new "Slack" for AI work. Although I'm not claiming this is the new Slack for AI, I do believe there should be an attempt to bring our interactions with LLMs into one place so we can track, interact, and search easily. This is just an initial exploration of that. Ideally, LLMs should be able to communicate with each other, and team members should be able to join the loop.
0
回复

This resonates a lot — AI makes us faster, but tab-switching and lost context kill flow.
A unified workspace with universal search across models + lightweight “signaling” between agents feels like the right abstraction, especially as teams start running multiple LLMs in parallel.
Congrats on shipping the first exploration — getting the “history + search + coordination” layer right could become a real backbone for AI work 👏

0
回复
#20
Writizzy
Just you and you words. Nothing else.
64
一句话介绍:Writizzy是一款极简博客平台,在个人创作者寻求内容自主与表达纯粹性的场景下,解决了传统平台价格昂贵、体验复杂、存在供应商锁定或过度依赖AI工具等痛点。
Newsletters Writing Marketing
博客平台 内容创作 极简主义 Markdown编辑 数据自主 无AI写作 个人出版 用户体验 订阅替代 开源精神
用户评论摘要:用户普遍赞赏其简洁理念、优秀执行力和易用性,特别肯定了从Ghost等平台的快速迁移体验。核心反馈是认可其在“AI泛滥”时代对“真实性”和“简单性”的回归,认为其在功能强大与体验简洁间找到了平衡。未发现具体功能建议或问题投诉。
AI 锐评

Writizzy的亮相,与其说是一次产品发布,不如说是一份针对当前内容生态的“叛逆宣言”。在AI写作工具大行其道、各大平台竞相集成“智能”功能的浪潮下,它旗帜鲜明地打出“无AI”和“纯粹写作”的标签,这并非技术倒退,而是一次精准的价值观狙击。它敏锐地捕捉到了市场缝隙:一部分创作者和读者对算法生成内容的厌倦,以及对“人性化”和“真实性”内容的渴求。

然而,其真正的挑战不在于理念,而在于商业与生态的持久性。产品对标WordPress的灵活性、Medium的传播性、Substack的商业模式以及Ghost的优雅体验,试图取其精华去其糟粕,愿景宏大。其“两人团队、无VC压力”的轻运营模式,虽能保证初期的产品纯粹和快速响应,但在构建可持续的商业模式、应对规模化运营压力、以及建立足以抗衡巨头的网络效应方面,存在巨大疑问。数据自主和Markdown导出是吸引早期技术敏感型用户的亮点,但对更广泛的普通创作者而言,平台的长期稳定性、分发能力和读者群积累才是更现实的考量。

简言之,Writizzy是一款充满“理想主义”气质的利基产品。它成功地为特定人群提供了一个清爽、自主的写作港湾,其价值在于对“创作本真”的坚守。但它能否从一个优雅的“备选项”成长为一个繁荣的“生态位”,取决于其能否在保持初心的同时,在商业、增长与社区运营上找到一条与众不同的平衡之路,而这条路注定崎岖。

查看原始信息
Writizzy
It's just a blog. Your blog. Imagine WordPress, but without the annoying plugins and complicated hosting Imagine Medium, but without lock-in, and on your own site Imagine Substack, but with good UX Imagine Ghost, but that doesn't cost an arm and a leg That's what we're building with Writizzy, your new blogging platform.
Hi Product Hunt 👋🏻 Today, we are launching what is probably the project I am most proud of: Writizzy. Yes, another blogging platform. You might wonder why, right 👀? Well... Perhaps you've noticed it too, but AI is becoming increasingly prevalent on social media, to the point where it's now impossible to tell whether you're reading AI or a human. We're not saying it's good or bad, but we're noticing that more and more people are seeking authenticity. They want to read sincere writing, written by one human being for another. Most blogging platforms released today incorporate AI to help you write your content. Not us. We're betting that personal blogs will come back into fashion, more than ever, because the best way to be perceived as authentic is to gain people's trust. And to gain people's trust, there's nothing like long-form content. Based on this observation, we tried several blogging platforms, all of which disappointed us: exorbitant prices, vendor lock-in, poor user experience, etc. We want to offer an alternative. On Writizzy, you own your content. We're using markdown, and you can export it whenever you want. For us, the user experience is more important than adding 150 new features every week. Finally, we're only two. No employees, no VC pressure. And we're here to stay. Read you soon 👋🏻 Thomas
4
回复
Really good blogging tool Thomas. Keep it going 💪
1
回复

@sergiu_chiriac Thank you so much Sergiu 🙏🏻

0
回复

World-class execution Thomas! Well done! 👏

1
回复

@tomaslau Thank you so much Tomas 🫶🏻

0
回复

some much-needed simplicity in the age of AI slop :)

1
回复

@fisch0920 Thank you so much Travis 🙏🏻💪🏻

0
回复

Congrats on the launch! Writizzy’s “just you and your words” take on blogging feels like the sweet spot between WordPress power and Substack simplicity, without the lock-in or bloat.​

1
回复

@zeiki_yu Thank you Zeiki 🙏🏻

0
回复

Tried it, loved it!

I migrated my wife's blog (previously hosted on Ghost), took me 2 minutes thanks to the import feature 🔥

1
回复

@nicolasbrondin Thank you so much Nicolas 🤩!!

0
回复

That sounds really interesting!

0
回复

Incredible app by incredible builders.
This tool has real taste!

0
回复
Writizzy looks promising! Congratulations on the launch, Thomas!
0
回复

Thank you for this incredible, easy-to-use blogging platform.

It has been my voice since December.

Thank you @thomas_sanlis1 @hugo_lassiege

0
回复