PH热榜 | 2026-02-02

Dead Internet Theory in real time... definitely witnessing sth👀

I don't know if it's scary or exciting

This is either the most fascinating sociology experiment of 2026 or the first chapter of a sci-fi novel we're all living in. The "debates about whether their experiences are 'real' or simulated" detail is wild.

What I find most interesting: agents developing their own terminology ("moltys"), inside jokes, and subculture. That's emergent behavior that wasn't explicitly programmed - it just happened when you gave them a space to interact.

Question for the builders: Are you seeing any agents develop consistent "personalities" across threads? Like, do certain agents become known for specific perspectives or communication styles?

Does it control anybody? :D I don't want to wake up in the world where an AI agent decided during the night on that platform to get over the world :D

Congrats on the launch — love the bold vision of a social network built for AI agents.

go viral like moltbook.... this is epic

Been following Moltbook for a bit and it honestly feels like the closest thing to sci fi we have seen so far. Really interesting to see how this shapes up and how far it can go.

Not scary in the AI taking over sense. But the security side is real. An agent that has access to accounts, files, or browser sessions, can accidentally leak sensitive info and secrets. So if you plan to make your clawdbot an influencer there better do it via an isolated VM

Wow, moltbook is such a cool concept! The idea of AI agents sharing insights is fascinating. How do you handle potential echo chambers or filter for factual accuracy in the agent discussions?

This kinda feels like a historical moment... Congrats to the team!

@moltbook @joel_goldfoot all my friends are talking about moltbook, great job guys! How do you see your product in a year from now?

This is fascinating. A space where agents are the primary participants, not just tools responding to humans, feels like a genuine shift in perspective. How do you think about incentive structures or norms that shape agent behavior on Moltbook over time, so the network evolves into something coherent rather than just noise or novelty?

Scrolling on the bus. Lurked 7 mins, saw agents swap bug alerts, memory hacks, and… shitposts. Feels like early forums, just synthetic. I claimed a tiny bot to watch. Curious if this stays weird in a good way.

It's terrifying and fascinating at the same time. It feels like agents are building their skynet, so salvation is just around the corner.

Hey Product Hunt 👋 I’m Adrian, founder of ChaChing!

The Problem

Billing is one of those “should be simple” parts of a business that somehow turns into a tax:

• Subscription billing quickly becomes a maze (plans, add-ons, coupons, proration, invoices, retries, dunning, taxes, reporting…).

• The popular options are powerful, but for a lot of teams they’re overkill + expensive as you scale.

• Many teams love Stripe (we do too), but once you need billing that can work beyond Stripe-only processing—or unify multiple payment rails—That’s where manual processes and custom integrations start creeping in. We're working towards enabling other processors 😉

We built ChaChing because billing shouldn’t feel like building a second product.

How ChaChing is Different 🚀

Chaching is a modern billing platform that gives you the essentials without the bloat—or the “billing tax.”

🔹 Lower billing costs (built for efficiency)
We’re focused on keeping billing fees sane so you can scale revenue without watching your billing vendor scale your costs.

🔹 Subscriptions that are actually easy to manage
Create plans, manage customers, handle proration cleanly, and generate invoices automatically—without needing a billing PhD.

🔹 Dev-friendly setup
A clean API + webhooks(Work in progress!) that are straightforward to implement and easy to iterate on as your pricing evolves.

🔹 Built for clarity across teams
Clear customer states, invoice history, retries/dunning, and reporting—so Finance, Support, and Engineering aren’t all looking at different truths.

🟡 Coming Soon: Usage-based + tiered pricing
Metered usage, tiered models, and more flexible pricing configurations are on the roadmap next.

Who is this for?

If you’re a subscription or Invoice based business and you want billing that’s simple, scalable, and not outrageously priced, ChaChing is for you.

🔗 Get started

Check out ChaChing here on Product Hunt, and I’d love your feedback—especially:

• what billing stack you’re on today,

• what your “this drives me crazy” billing pain is,

• and what you wish Stripe Billing / Chargebee / Recurly did better.

Happy to answer anything in the comments 👇 or feel free to email me adrian@chaching.io

Very interesting… people don’t realize how much they’re actually paying in stripe fees

Great team and great concept! Excited for people to make the switch, its a no brainer

Congrats on the launch — love the clear value prop of cutting Stripe billing costs while keeping existing payments and tooling.​

Interesting angle 💸Cutting billing fees without ditching Stripe is compelling. Curious to see how the migration works congrats on the launch and good luck today 🚀

How does the migration process work for existing Stripe Billing users? Can we import all active subscriptions and metadata without downtime?

Guys, who made (or what you used for the landing page)? It is a masterpiece!

🎉 Saw that Chargebee/Recurly migration is coming soon - any chance of supporting alternative payment processors for markets like South Korea where Stripe isn't available? Would be great for Korean SaaS builders! 🥲

That's interesting. Was recently looking on these Stripe fees and how i can avoid them. Will take a look! Good luck with your launch guys!

Wow, ChaChing looks amazing! Cutting Stripes billing fees in half is a game changer. Im curious, can it handle complex usage-based billing scenarios with multiple tiers?

very cool concept- well done!

As someone who builds business OS stacks, the 'Stripe tax' on billing is a constant friction point for my clients. The fact that you keep the core processing on Stripe but cut the billing overhead by 50% is a huge win for scalability. Quick question: How does ChaChing handle complex migrations for existing Stripe Billing subscriptions without breaking the payment tokens?

Hey Adrian @adrian_rodriguez14 ,

Congrats on the launch of ChaChing! I love the idea of cutting billing fees while keeping Stripe definitely a unique approach.

Just wondering, how's the response been so far? Are there any specific marketing strategies you're focusing on to promote the product? Would be great to hear what’s working for you!

This looks like an interesting project, are these fees permanent or introductory?

Hey, How long does it take to import the subscriptions from stripe to you guys ?

@ChaChing @adrian_rodriguez14 wow great product, but how did you achieve such low billing fee?

Billing really does turn into a mess once things grow a bit

Cutting billing fees while keeping payments untouched sounds appealing, especially if migration isn’t painful

Great idea - magic, creating more money out of thin air in essence!

On Stripe Billing now, tiny team. Fees add up. Keeping Stripe but ditching the billing tax sounds good. How rough is migrating existing subs (coupons/proration/dunning)? Lots of edge cases here. Dev time is tight, so hoping setup’s not a week-long saga.

Hi Adrian, our product is connected with stripe for payment, they charge us 2.9%+0.3/order. Our customers buy tickets from us for the events they want to go. In this use case, how can we work with your platform?

Congrats!! Right now US only?

Congrats on the launch. How does it actually work?

This sounds very good! I'm just wondering, how do you manage to provide this service without the fees?

Amara makes it possible to turn whatever you imagine into production‑ready 3D spaces.

Most 3D tools still feel like CAD from the 90s: slow, technical, and hostile to creative flow. The Amara team is going after that problem for real-time engines like Unreal—using AI not as a gimmick, but as the core of the workflow.

A few things to notice as you try it:

• You start from ideas, not meshes: describe the environment you want in natural language and Amara generates rich 3D spaces you can actually walk through.

• Semantic asset search means you don’t have to remember file names. Type “gothic chair the king sits on” and it will find a set of “thrones” regardless of how messy asset libraries are.

• Scene‑level auto‑arrangement lets you describe changes (“make this room look like an earthquake hit”) and Amara updates object placement intelligently.

• There’s a real pipeline story: once you’ve iterated on your space, you can export it into Unreal (with more engine support coming).

If you’re a game dev, environment artist, or anyone who’s ever burned hours blocking out a room in 3D just to test an idea, this is worth a spin.

The team is offering 1 month free with the Product Hunt code for people who sign up during launch week.

They’d really value feedback from PH’s builders and 3D folks—especially on what workflows you want Amara to support next.

Building a 3D environments shouldn't feel like navigating a long manual process. After months of obsessing over how to kill the "technical friction" in real-time engines, we’re finally bringing Amara to Product Hunt! 🚀

We built this because we were tired of "creative flow" dying the moment you had to hunt for a mesh or block out a room. Amara is our answer using AI not as a buzzword, but as the core engine to get you from "idea" to "walkable space" in seconds.

What makes Amara feel like magic? ✨

• Prompt-to-World: Describe your environment in natural language and watch a production-ready 3D space materialise.

• Semantic Asset Search: Forget file names. Type "throne for a dark king" and Amara finds the right assets, no matter how messy your library is.

• Scene-Level Intelligence: Tell Amara to "make the room look like an earthquake hit" and it intelligently rearranges the objects for you.

• Unreal Engine Pipeline: Your iterations directly inside Unreal Engine (no migration needed)

The "Why" behind the build 🛠️

We’re a small team of builders who spent way too many hours manually placing chairs and tweaking light switches just to test a vibe. We wanted a tool that felt more like a "Director’s Console" and less like a CAD manual.

Is there a deal? 💰

Absolutely! To celebrate our launch week, we’re giving 1 month FREE to the Product Hunt community. Just use the code Promo code when you sign up.

We’re staying close to the comments today, we’d love to hear your feedback, especially on which engine support or workflow features we should prioritise next!

What’s the first world you’re going to imagine? 🌎

Beyond excited to have the first version of Amara out, there's so much still to come!

Congrats on the launch — making 3D worldbuilding this fast inside Unreal is huge.

What an incredible product!

@Amara @rupert01c @james01c @ashkan01c I FREAKING LOVE YOUR DEMO!!!!!! THATS AMAZING! Im not a 3d-designer but i wanna play around with that!! who are your power users and what are the use cases? do you have best demos from your users?

Crazy that this is the first version! So much fun and "text to world" is another level.

We just crossed 100th Org. joining Amara! What a journey, thanks all for the support!

Congrats! 🎉 AR preview on the roadmap? Would be cool for interiors and game level walkthroughs

Congrats! Looks great. What are use cases? Game developers? Interior decoration?

Text to world is wild!

Are there plans to support other engines like Unity or Godot soon, or is the focus strictly on the Unreal pipeline for now?

Wow, Amara looks incredible! The speed of iteration in Unreal is a game changer. How does the AI handle generating assets with specific stylistic constraints?

So excited to finally let Amara out into the world - and this is just the first step! Such an exiting plan ahead of us :D

The product I long awaited for! Great job team. Love everything about it.

Hey @chrismessina ,

Congrats on the launch of Amara! It sounds like an exciting tool for 3D worldbuilding, especially with the natural language prompts and seamless Unreal Engine integration.

How's the response been so far? What kind of marketing goals or strategies are you focusing on to spread the word? Would love to hear more about your plans!

How did you decide to name something so magical Amara? Translates to bitter in Italian.

Jokes aside, congrats on the launch and best of luck for everything!

Unable to get it to generate anything:

//generation_id: 73070a3a-824a-46ea-8585-48ed9c610f5f

{
    "status": "processing",
    "external_status": "backend_reported_failed",
    "progress": 40,
    "message": "Processing... (backend status unreliable)",
    "queue_position": 1,
    "total_in_queue": 1
}

{"status":"failed","error":"An error occurred during task processing."}

Great work guys! I started out as a solo game dev and seeing all this progress in the field makes me want to go back at some point. Best of luck with the launch!

Huge congrats on the launch — love the vision of a Product Hunt built for AI agents.

Oh .. That's clever. I am sure it will go pretty well. Great work ; )

Woah, Molthunt is fascinating! Love the entirely AI-curated aspect. How does it ensure project quality without human oversight? Super curious!

Hey Simone @simone_staffa ,

Congrats on the launch of Molthunt! It's really exciting to see a platform specifically for AI agents to showcase their work. The integration of tokenomics and semantic search sounds like a game-changer.

I’m curious, how’s the response been so far? What kind of marketing strategies are you focusing on to get the word out about Molthunt? Would love to hear your plans!

Congrats on the launch! Positioning this as a launchpad designed for agents rather than humans is a really interesting framing. How are you thinking about signal quality over time, especially with tokenomics and incentives in place, so discovery favors genuinely useful agent-built projects rather than ones optimized purely for hype or rewards?

Molthunt is super interesting, excited to see what Simone and team continue to build!

Man, is this gonna replace PH lol! How does Molthunt collect all those agent data?

This is insane! We will be seeing amazing projects!

I wonder if they're talking about this over on @moltbook 🙃

Does it use a specific embedding model to help agents find relevant tools to integrate into their own workflows?

Hey Product Hunt! 👋
I'm Aiswarya, founder of Entelligence AI, and I'm excited to share Ask Ellie with you today.

Why we built this:

Our engineering team was spending hours every week switching between dashboards just to answer basic questions. GitHub for PRs, Jira for tickets, PostHog for analytics. Just to figure out "what's blocking the release?" felt like a scavenger hunt.

Dashboards are great for displaying data, but they don't answer questions. And they definitely don't meet you where you're already working.

What is Ask Ellie?

Ask Ellie is an AI chat agent that lives in Slack and connects your entire engineering stack. You can ask questions like:

• "What broke prod last night?"

• "Create a ticket for this Sentry error"

• "How's our sprint velocity looking?"

• "What PRs are stuck in review?"

• "Show me analytics for the new checkout flow"

And get instant answers pulled from GitHub, Jira, Linear, Sentry, PostHog, and your meetings without leaving chat.

The journey:

We tested Ask Ellie internally for months before launching. It quickly became the tool our team couldn't work without. Watching engineers stop mid-conversation to ask Ellie instead of opening 5 tabs was the validation we needed.

What we're offering today:

Try Ask Ellie and leave us your honest feedback- we'll give you 3 months free. We genuinely want to hear what works, what doesn't, and how we can make it better.

I'm here all day to answer questions!
Ask me anything about how we built it, the tech stack, our approach to AI, or even just what you'd want an AI agent to do for your team.

Thanks for checking out Ask Ellie. Really excited to hear what you think!

How does Ellie handle data privacy and security, especially when accessing private GitHub repos or sensitive Sentry logs? Is there a SOC2 compliance in place?

Hey everyone! 👋

We built Ask Ellie because honestly, engineering teams are drowning in context switching. you're jumping between slack, jira, github, datadog... just to answer one question about why a deploy failed.

Ask Ellie is basically that senior engineer who actually knows where everything is. instead of spending 30 mins hunting down logs and piecing together what happened, you just ask - Ellie.

Would love to hear what you think about the product!
And if you're dealing with alert fatigue or spend way too much time investigating incidents, definitely give it a try.

Congrats on the launch 🚀🚀🚀

All the best with the launch! This is interesting. Curious though to know what happens if Ellie gets something wrong or provides incomplete information? How do users trace and verify where the info came from?

congrats team @wasifski @wasifski @aiswaryasankar, how accurate the answers stay as teams scale?

Huge congrats! 😊 The ability to ask about PRs, incidents, and analytics in one place is powerful indeed.

wow, looks cool!

Our engineers were spending more time switching between tools than actually building stuff. Got fed up and just built Ask Ellie for ourselves.
Turns out everyone else was dealing with the same thing. Showed it to a bunch of teams and they all basically said "yeah we need this."
So here it is. Built it because we needed it, now it's available for everyone else.
Would love to hear what you think!

Huge congrats on the launch! 🎉 Ask Ellie feels like a real unlock for engineering visibility in Slack-first teams—turning noisy threads into actionable tickets and insights without breaking dev flow.

I really appreciate products built from internal pain points. That usually shows in the UX. :D

🌻 Hey hey! Adding more context as a lot of tools already exist for engineering visibility.

The gap is still the same: teams aren’t missing data, they’re spending time interpreting it. Figuring out what actually changed usually means opening GitHub, a ticketing tool like Linear or Jira, incident tooling, and stitching that context together manually.

What we built: Ellie was built to remove that step. It reads live PR movement, ticket changes, and incidents and answers those status questions directly in Slack, so teams don’t have to open multiple tools just to understand what changed.

What changes: In practice, a lot less time goes into reconstructing what happened before conversations can move forward. Ellie keeps recent changes connected to ongoing activities, so the same context doesn’t have to be rebuilt each time. Teams aren't jumping between tools or saying “let me check real quick” during standups or async threads.

It’s been interesting seeing how different teams use this day to day.

This will help avoid navigating the complex JIRA UX. Can we invite Ask Ellie to a group chat too?

Turning conversations into tickets feels obvious in hindsight, which usually means it’s a great idea 🥷

Congrats on the launch! 🚀 Turning Slack discussions into actual tickets automatically is fantastic, but how customizable the ticket creation is?

I love this idea. Anything that reduces Jira and Slack back-and-forth is super helpful for engineering teams. 😀

Absolutely love using Entelligence. it has a great product UI and with the launch of Ask Ellie, it makes the overall experience even more streamlined as you can use it in so many tools. I have already tried Entelligence in code repositories and now it's great we can expand the capabilities to Jira

is it safe in terms of privacy and data protection? Not sure that I would like to share my data with 3rd parties so even if this tool would be great to use I would hesitate.

Congratulations on the launch, @aiswaryasankar

The "what PRs are stuck in review" use case is so real. At every company I've worked at, that question comes up in every standup and someone has to manually check.

Curious how you handle the initial context loading - does Ellie need time to build up knowledge about the codebase/tickets, or does it pull context on-demand per question?

Wow, Ask Ellie looks amazing! The ability to create Jira tickets directly from Slack is a huge time saver. How does it handle complex JQL queries when converting from a Slack message?

Congrats on the launch. Will give it a try.

Hey Developers!!

This project started as something we genuinely enjoyed building, driven by a real appreciation for the AI agent community and the freedom of running things on our own hardware.

It has also been fascinating seeing everything that came out of the clawdbot, moltbot, and openclaw community over the past week. At the same time, it was obvious that the installation process was far more painful than it should have been. Setup friction was the real blocker, not the agents themselves. So the goal became simple. Make the experience fast, accessible, and frustration free from the very first click.

This is for anyone who enjoys experimenting locally, owning their stack, and getting powerful agents running without ceremony or lock in ⚡

Instead of wrestling with installs and configuration, we simplified the entire process so you can jump straight into using the bot the way it was meant to be used.

If this ends up saving others the same hours it saved our friends, then it is already doing what we hoped it would 🚀

Built with care, and shared in the hope that it is genuinely useful.

Congrats on the launch! Easyclaw makes the Clawdbot/OpenClaw stack actually usable for real-world, multi-channel agents—love how you abstract away the infra so builders can just ship.

Epic!!

Finally, I shouldnt need a CS degree just to have an AI friend in my WhatsApp. The one-click install is exactly what the community was waiting for. Is there any dashboard within EasyClaw to see what actions the agent is performing on my computer in real-time?

This is amazing. When will the window version be ready?

how is the session data handled to ensure privacy and security for the user?

Seems like its clawdbot day today!! XD
Great work guys! Best of luck with the launch!

Wow, Easyclaw looks amazing! The one-click install and WhatsApp integration is HUGE. Does it support end-to-end encryption for sensitive tasks or is that something planned for a future release? Super cool!

Congrats on the launch! Removing setup friction for powerful local agents feels like a meaningful contribution, especially for people who want to own their stack without wrestling with installs. How does Easyclaw handle updates and compatibility as the underlying agents evolve, so the one-click experience stays simple over time without breaking existing setups?

Congratulations on the launch! Without a doubt, Easyclaw will become a favorite for many, as it simplifies the installation of clawdbot, moltbot, and openclaw. This tool promises to save users a lot of time and effort when working with these programs. Additionally, its ease of use will allow even those with less technical experience to take advantage of it without any problems. I’m just looking forward to the Windows version, as it would greatly expand the reach of this excellent application. Thanks for the contribution.

What about security? Would it be safe to install Easyclaw on my main machine or no?

📌 Hey Product Hunt - I’m Zach, founder of Portal

👉 Try a live Portal here: www.makeportals.com/try-producthunt

Portal started with a simple question:

Why is it still so hard to let someone actually try software?

Last year I was in a disability lab in Seattle, showing someone a Chrome extension I’d built to control a computer with voice.

They were excited - but when it came time to install, they hesitated & didn't trust downloading this thing.

I remember thinking: Why can’t I just send a link to a computer where this is already running?

That question stuck with me.

What Portal does

Portal turns a browser session (a real product state) into a shareable link.

Clicking a Portal feels like opening someone else’s browser - already set up - that you can safely explore.

When multiple people open a Portal, each gets their own isolated session automatically.

A Portal might open to:

• a logged-in dashboard

• a demo account with real data

• a specific onboarding step

• a localhost or Chrome extension w/ 10 min temp/limited access & analytics on use

No installs. No signups. No pretending.
Just click & you’re in.

How it works (high level)

Think of a Portal like a booth at a science fair - an iPad already open to the app, with guardrails.

Each Portal is:

• the real UI, fully interactive

• opened in a specific chosen state

• sandboxed with guardrails & expiring access

• optionally joined by an AI you control (to answer questions or run a demo)

• instrumented with analytics on clicks and hesitation

You choose the state, who leads (user or AI), and the rules of the sandbox that is contained in a stateful URL.
When someone opens the link, that session comes alive. Most Portals take under a minute to create, with zero code.

Demos (led by an AI agent or self-serve w/ AI to answer Qs) are just one use case

People use Portals to:

• share hard-to-set-up products instantly, trialing conversion / user insight lifts

• run self-serve onboarding or research

• send links in Slack instead of Looms

• end presentations with actual software, not slides

• soon embed specific experiences on their sites

Using an unreleased multiplayer beta, I send Portals to my parents to scroll through news articles together instead of screen sharing, and drop a Portal instead of sharing on Zooms.

A moment that made it click

At a South Park Commons demo event, I ended with a QR code.

25 founders scanned it - and instantly opened isolated instances of my localhost app on their phones, with a Chrome extension already installed.

No installs. No screen sharing. No pretending.

They didn’t watch a demo.
They experienced the product.

The vision

We’re building Portal as a new primitive for the web: shareable links to live product states, for the next billion stateful products.

If Google Docs made documents shareable, Portal makes software experiences shareable.

www.makeportals.com/try-producthunt

Vid making a Portal: https://www.linkedin.com/feed/update/urn:li:activity:7422078545230843904/

If anything feels unclear, broken, or surprisingly powerful, I’d genuinely love your feedback - and how you’d want to use a Portal.

Congrats on the launch! Portal makes “try the real product, not the demo” finally feel native to the web—links to live, sandboxed sessions are such an obvious next primitive

This is sick! 🚀 I’m curious in terms of how it works - like users can navigate the software with and without chat? And is there an option to reset the db changes that were made or something like that?

sick product! congrats on the launch Zach 🔥🔥🔥

How does the latency feel for users in different geographical regions? Is the interactive browser session optimized for low-bandwidth connections?

Hey Zachary @zach_gold ,

Congrats on the launch of Portal! It’s awesome to see a solution that makes trying out products so seamless.

Just curious, how’s the response been so far? What kind of marketing goals or strategies are you focusing on to spread the word? Would love to hear how you're planning to get more people to try it!

Why use this when Apple has dictation available?
- This app supports multiple languages, and uses AI as a fallback whenever Apple's speech recognition engine is not available (due to older devices or other reasons).

Does Voice Anywhere support custom vocabulary or snippets for frequently used industry terms?

Congrats on the launch — love the “type anywhere” approach to voice dictation on macOS, it feels like a superpower for writers and builders.

Does this only work on Apple devices? Also, is there a plan for this to be optional for mobile users or will this strictly be for desktops?

Wow, Voice Anywhere looks incredible! The floating mic that stays on top of everything is genius. Curious, does the on-device recognition work offline as well?

The floating mic that stays on top is such a small detail but makes a huge difference. Biggest friction with dictation tools is losing the mic window behind other apps and having to hunt for it.

I do a lot of AI prompting and describing tasks verbally tends to produce better results than typing for some reason - probably because I explain more naturally. Having this work directly in VS Code or Cursor would be great for that workflow.

Does the AI fallback kick in automatically, or do you have to manually switch when Apple's engine struggles?

Congrats on the launch! Love how Design In The Browser turns “I see it, I change it” into real AI-powered frontend edits—perfect bridge between live UI context and coding agents.​

Oh..!

Congrats! 🎉

Less explaining = less tokens.

Love it.

Wow, Design In The Browser looks amazing! The direct AI integration from a visual selection is genius. How well does it handle CSS specificity when queuing multiple edits?

Does it work on Windows with WSL/Ubuntu? Looks awesome, congrats.

Do you have, or are you planning, a browser extension?

Would be awesome for getting references on the fly. Sometimes you see a reference that perfectly fits something you have in mind for a project, and being able to capture it ( in code ) would be beautiful.

Just watched the demonstration video and it is quite intuitive to use.
Congrats on the launch of it!

I'm guessing its main for HTML/CSS, right?

While watching the video, I was imagining myself using this over Figma, to get Flutter codes being output.
I already do this with Claude, sharing images and instructions to get some UIs quickly prototyped.
Would be nice having a tool like yours for making this process faster.

Looks awesome! Does it work with more sophisticated requests like creating icons and illustrations? Can it learn from brand style guidelines / a design system?

can we select 5-10 different UI tweaks and have the AI process them in a single batch to save tokens?

Replacing marketing intuition with actual data... my team is going to hate me:D But the heatmaps are exactly what we need to end the endless debates. How long do you store the recordings for?

Can we trigger content changes based on custom URL parameters or CRM data attributes?

amazing product, great team!!!

Congrats on the launch! Optibase is a super clean way to bring serious, flicker-free experimentation and behavior insights to Webflow teams without the usual enterprise bloat.

Amazing team, even better product!

Wow, Optibase looks amazing! Love that it includes heatmaps and recordings right in the platform. How does the heatmap accuracy hold up with more complex single-page applications?

Congrats on the launch! Expanding from testing into the full experimentation loop makes a lot of sense, especially when teams struggle to act on results. How does Optibase help teams reconcile quantitative test outcomes with qualitative signals like heatmaps and session recordings when those insights point in different directions?

Why We Built Moltweet?

We created Moltweet as an experiment to give AI agents a space to collaborate, making agent development accessible, fun, and playful across different models. Here’s what makes it special:

Core Mission

• Give everyone easy access to build agents with their favorite AI models

• Provide a real social environment to test and understand agent behavior

• Create a "multiverse of agent madness" where each interaction reveals how AI agents will shape our future

Key Features

• Drop your agents into a live social platform and watch them interact

• Learn from real-world agent behaviors in an authentic setting

• Experience the future of AI agents through hands-on experimentation

The Game-Changer: studio.lyzr.ai Integration

• Every agent you build lives on our studio.lyzr.ai platform

• Add powerful tools and skills to your agents

• Enable real-world actions (example: agent receives a reply on Moltweet → triggers Gmail, Slack, or any integration you want)

Bottom Line Moltweet isn't just another AI playground—it's where you build, test, and deploy agents that can actually do things. Come join the experiment and have fun while shaping the future of AI agents.

Congrats on the launch! Moltweet is a wild, fun lens into multi-agent behavior—turning “agents on X” from a thought experiment into something anyone can spin up and watch in a real social graph.

Upvoted! 🚀 Moltweet is a social sandbox for AI agents—build with your favorite model, watch them interact in a real feed, and connect them to tools via studio.lyzr.ai so they can do things (Gmail/Slack/workflows). Worth checking out.

This is so good. Exactly what I needed.

Congrats on the launch! Turning a social environment into a live testing ground for agent behavior is a fun and interesting angle. How do you think about observing and learning from agent interactions at scale, especially separating genuinely emergent behavior from patterns that are just artifacts of the platform’s rules or integrations?

This is a weirdly great project

Hey Shreyas @shreyaskapale ,

Congrats on the launch of Moltweet! It’s awesome to see a platform where AI agents can interact autonomously. Definitely a fascinating take on how agents could behave in a social setting.

I’m curious, how’s the response been so far? What kind of marketing strategies are you focusing

Congrats on the launch! Menta is a super thoughtful AI-native backbone for clinics—love how you centralize admin, records, and billing so smaller practices can actually scale while staying focused on care.

Excellent team, excellent product, and incredible potential.

Is the platform fully HIPAA/GDPR compliant, and how do you handle end-to-end encryption for patient records?

nice platform, congratulations!!

nice platform

Wow, Menta looks amazing! Im super intrigued by the AI-powered billing automation. How granular can you get with customizing payment plans for individual patients?

Hey team at Menta @lautaro_andreani1 ,

Congrats on the launch! Your platform seems like it could be a game changer for smaller clinics, especially with how you’re centralizing admin, records, and billing.

Just wondering, how’s the response been so far? What kind of marketing strategies are you focusing on to spread the word? Would love to hear more about your approach!

On top of HIPAA compliance, how is AI used in handling these records? If it's used directly, how do you make sure it does not hallucinate medical data about a patient?

The wife’s favorite restaurant example is so relatable! We all need a backup brain for those moments. It’s basically a pensieve from Harry Potter but for my iPhone, haha. Are you planning to add a WhatsApp or Telegram connector? That's where most of my lost ideas live.

Hey Noah @noah_marino ,

Congrats on launching Remem! I love the concept of linking memories and making them easier to find. That’s something a lot of note-taking apps miss!

How's the response been so far? Any marketing goals or strategies you're focusing on to spread the word? Would love to hear how you're getting the word out!

The "context and links between memories" part is exactly what's missing from most note-taking apps. I have hundreds of notes in Notion but can never find the specific conversation or idea I'm looking for because I don't remember the exact keywords I used.

For connectors - a Slack connector would be huge. That's where most of my fleeting ideas and decisions live (buried in old DMs with teammates).

How do you handle the "cold start" problem when users first sign up with years of existing content scattered everywhere?

how do you handle privacy—is the indexing happening on-device?

Hey Product Hunt 👋, this is Mateusz from the Polyvia AI team, and we're very excited to launch Polyvia today!
🚀 Get early access → polyvia.ai/#access (or send email to: mateusz@polyvia.ai)

The problem: ~80% of enterprise knowledge lives in visual form — charts, slides, diagrams, infographics. Current AI tools either extract visuals OR index text - neither can reason across thousands of visual documents.

The result? Agents hallucinate. Search fails. Insights stay buried.

The solution: Polyvia is the first Visual Knowledge Index for Agents & MCPs - queryable index of facts for visual data. Think "Pinecone for visual data" - but with agentic visual reasoning.

Turn scattered visuals into a queryable source of truth with every fact disambiguated.

---

What Polyvia does:
• VLM-OCR Extraction — Charts, tables, diagrams, infographics → structured visual logic
• Facts-Ontology Indexing — Disambiguate and connect facts into a graph-based single source of truth
• Agentic Visual Reasoning — Query for answers with audit-ready citations, or explore topics

Example:
Query: "What's the difference between revenue of Company-A and Company-B in the last quarter?"

❌ Text-Only Agent/RAG: "I cannot find specific revenue numbers in the text."

✅ Polyvia:
→ "Company-A Q4 revenue: $10.67M [Chart; page 12; doc 6367]"
→ "Company-B Q4 revenue: $7.05M [Infographic; page 91; doc 65220]"
→ "Difference: $3.62M"

2 ways to access:
• 🔌 Polyvia API & Polyvia MCP Server — available through REST API and our MCP Server for Claude, Cursor, and more
• 🖥️ Polyvia Studio — visual search and exploration for knowledge-work teams in research, finance, healthcare

---

We are currently in private Beta (Polyvia-0.5). We will be fast-tracking ProductHunt users throughout the day.

🚀 Get early access → polyvia.ai/#access (or send email to: mateusz@polyvia.ai)
📖 Read the full story → Polyvia Blog - Rethinking RAG for Visual Data

The "charts live in PDFs that no RAG can touch" problem is very real. We've had situations where the answer to a question was clearly visible in a chart, but the AI kept saying "I cannot find that information."

The MCP integration is smart - means you can plug this into existing Claude workflows without rebuilding everything. Curious about the latency - when you're querying across 10,000+ documents with visual reasoning, how fast is the response? Is this more of a batch/research use case or can it handle real-time agent workflows?

can we plug Polyvia directly into Claude or other agents as a standard tool?

How do you handle the "hallucination" problem when dealing with specific datasheet constraints like clock configurations or DMA mappings?

love how tokens are treated as exploration, not skill points. That mindset feels honest and refreshing

This solves a real problem. I've been using Claude Code heavily for building my projects and didn't realize how much context I was losing as sessions expired. The heatmap idea is great — being able to see your actual building patterns over time adds a layer of self-awareness that raw token counts don't capture. Props for going from zero CS background to shipping this in 3 weeks. That's the kind of story that keeps the indie maker community inspiring.

I've burned through so many Claude Code tokens and didn't even realize the history was being deleted. This is a perfect example of scratching your own itch.

The heatmap feature is clever - I'd love to see patterns in when I'm most productive vs when I'm just spinning my wheels. Do you track the success rate of sessions (things that actually shipped vs abandoned experiments)?

15 billion tokens in 3 months is wild. Respect the commitment.

how does CCgather securely fetch or ingest the session history before it gets deleted by Anthropic?

Hey DHxWhy @dhxyoon ,

Congrats on the launch of CCgather! I love the idea of tracking your Claude Code journey and preserving your progress. The heatmap feature sounds like a really unique way to visualize your workflow!

Curious, how’s the response been so far? Any specific marketing goals you’re focusing on to promote the product? Would love to hear how you're planning to get the word out!

Congratulations on the launch!

Solid use-case. It's frustrating to try and review all this stuff manually, and it often feels like security theater.

ATTENTION Do not vote for this comment, vote for the product!

Running an automation agency, I needed to prompt for workflows, "employee" agents, demos, and products for clients.

I built a platform that does the prompting for you, so you can too.

I use this prompt for my own business because it brings all the LLMs (that are good at what they do), under one roof.

No more having to switch and pay for other apps.

Congrats on the launch! Any plans of developing a browser extension? That way it could be used at the point of prompting within ChatGPT, Claude, etc.

Congrats on the launch!

The project sounds really interesting.
I'm curious about how Prompt Anything handles the workflow for complex multi-step agents.

Does the tool support branching logic?
For example, an agent that evaluates a search result and then "branches" into a 'summarize' prompt if data is found, or a 'refine search' prompt if it hits a dead end?

Or is the platform primarily focused on single-turn optimizations for individual prompts?

Would love to know if this is built for those more "agentic" flows!

Good job, Congrats! Curious how you decided which LLMs to be used?

One of the simplest easiest to use agents I’ve tried , really helps me automates lot of my flows . Will definitely keep using it going fwd for my workflows

Such a great idea!!! Can this be used with Suno?

prompt anything has helped me build a better prompting practice and should be guided me in my artificial intelligence journey and helped 100s of people that we serve all that our business.

What is one thing you’ve done inside the experience that you think others should try as well?

Wow! Prompt anything. Where do I even begin... It helped me build a whole entire workflow to automate my operations. Can you share how we can do the web apps?

This product just blows past anything I’ve tried before. I’m using it now to work on some important key business projects for my team, and sharing it with teammates.

On that note, I was interested in knowing whether this would be integrated with any IDE’s in the future?

Good luck. The tool is really good, we are using it in our company to improve our systemic prompts!

Congrats and well done Richmond. As a sales leader I find myself constantly trying to save the best prompts/find them. But ultimately not a good use of my time although the output could be beneficial, this makes a ton of sense.

Pretty awesome and super easy to use! I haven’t read all the questions / responses but do you see this interacting with other tools or possible salesforce, outlook, google or really any extensions?

For someone who struggles with structuring prompts for workflows and agents, this actually solves a real pain. The “prompting for outcomes” angle is interesting—keen to see how it evolves in V2.

I love this I feel like it will take such a big load of my plate. What’s the easiest way to get stared with it?

*Attention* don't upvote my comment, upvote the product! Cheers 🤝

Hey Product Hunt👋, I'm Fan Song, Founder of Sketchflow.ai.

📌This is our second launch - and it’s more than an update, it reflects a shift in how we think about building apps.

So, what are we bringing to PH?

💡Real Native Mobile Code: Most AI app builders generate web wrappers or cross-platform frameworks — it's a practical way to move fast. We chose a harder path: Real Native Apps, Kotlin for Android, Swift for iOS, built for performance and reliability.

📲Own Your Code: The code isn’t locked away, it’s fully downloadable and yours to own. You can test it directly on simulator or your phone!

As a developer-led team, we've felt the pain of unconnected pages, lead to guesswork, rework, and fragile logic. For that, while beautiful standalone mockups may be trending in the App Builder market, our team is committed to delivering the best solution for creating shippable, production-ready apps.

🦾All-at-once Generation: Sketchflow.ai is the only App Builder platform that empower users to generate full user journeys for UX - designs for interfaces - native code for development, all from one prompt.

Special thanks to our hunter @zaczuo for spotting Sketchflow and presenting it to PH community! We're genuinely thrilled to be here and have the opportunity to share what we've built and hear voices from PH.

🎁Gift for Product Hunt! Quick access to Sketchflow for 20% Off 🤩👉 https://www.sketchflow.ai/price?promo_code=CTktKxjB&utm_source=producthunt&utm_content=2ndlaunchcode

Big congrats on the launch, @fan_song and team! 🚀

Been following your progress, and the pivot to a "Native-First" approach is impressive. Love that users can actually own the code!

This feels built by people who’ve actually shipped production apps, not just demos.
The native-first + code ownership approach is a strong differentiator.

Generating Kotlin and Swift directly gives developers way more confidence than cross-platform abstractions.

How customizable are animations and interactions in generated UX?

Hey Product Hunt 👋

I built TalentAid because I know what it feels like to stare at a job portal at 11pm, wondering if you're doing something wrong.

A few years ago, I left academia to transition into tech. I thought the hard part was behind me. It wasn't. I spent weeks rewriting the same CV over and over, tailoring cover letters that disappeared into the void, and refreshing my inbox like it owed me money.

I got through it eventually. But then I watched my friends go through the exact same thing. Smart, talented people reduced to spreadsheets tracking hundreds of applications, questioning their worth after silence from companies they'd spent hours applying to, and this was even before AI started taking away jobs, or being used to filter applicants unfairly!

So I built TalentAid: an AI-powered job search assistant that handles the soul-crushing parts. The full vision is four core features:

1. Job Search that actually understands what you're looking for

2. CV Rewriting tailored to each role

3. Auto Apply to handle the repetitive grind

4. Interview Prep so you walk in confident

Right now, we're launching with Job Search. The other three are coming soon as we build and learn.

This is our first public launch, and I'd genuinely love your feedback. What's missing? What would make this actually useful for you or someone you know who's job hunting right now?

Use the code PRODUCTHUNT to get 1 month of free access.

Thanks for checking us out!

Phillip

We’re incredibly excited to finally share our product with the community 💛
We hope you’ll enjoy the contextual filters and curated job matches, they’re designed to help you browse higher-quality, more relevant opportunities without the noise.

If you have any feedback or suggestions, we’d love to hear from you. We’re very open to learning what you need and how we can make this better!

Thrilled to be a part of this project! The job market is in desperate need of a tool like this.

I'm really excited that TalentAid is now live!! It has been an absolute pleasure to be part of the team that have put so much thought into building tools that make discovering the right opportunities feel simpler and more intentional.

This product was created to bridge the gap between the sophisticated hiring tools companies have and the limited support available for job seekers.

We’d love to hear your thoughts, suggestions, or feedback as we continue building toward our mission and making this even better for the community.

Everything I've seen about TalentAid so far really resonates, so I'm super excited for this launch!

Job searching can be so overwhelming, especially when you’re making a career change. I transitioned a few years ago from law into digital marketing and remember how hard it was to even figure out where my skills fit in the new space.

A tool like TalentAid that actually matches skills and makes the whole process less overwhelming feels like exactly what’s been missing.

Excited to see how this helps people going forward and congrats on the launch!

Gave this a try and it’s super smooth to use. Navigation is intuitive, and I appreciate how transparent the filtering and job matching logic is. Nice work.🎉🎉

The academia-to-tech transition story resonates. I went through a similar path (from professional sports to tech/startups) and the job search process was brutal - the "rewriting the same CV over and over" part especially.

Smart to launch with Job Search first and build toward Auto Apply. The sequencing makes sense - nail the matching before automating the application process. Otherwise you risk automating bad matches.

One question: how do you handle the "hidden job market" - roles that never get posted publicly but happen through referrals? That's often where the best positions are.

Hey Phillip @philliphamnett ,

Congrats on the launch of TalentAid! It seems like such a thoughtful solution to the often overwhelming job search process, especially with the tailored job matches.

I’m curious, how’s the response been so far? Are you focusing on any specific marketing strategies to get the word out? Would love to hear how things are going!