PH热榜 | 2026-04-15

Hey Product Hunt 👋

We've been here before. In 2021, you made Fathom #1 Product of the Day. In 2023, Fathom 2.0. Today we're back with something that feels genuinely different and we think you'll feel it too.

Two years ago, a great AI notetaker meant one solid summary when the call ended. That was the bar. We cleared it, and kept building.

Here's what's new:

🎙️ Bot-free capture - your call, your way The #1 feedback we've heard for years: "the bot joining my call is awkward." We heard you.

• Choose meeting by meeting: bot-free transcript-only or audio capture (my personal favorite), or full video with the bot

• All three deliver the same high-quality summaries, action items, and full speaker attribution

• Mute-detection built in - we respect your privacy

• Works in Slack Huddles too, so your quick syncs are as well-documented as your scheduled calls

🤖 Claude + ChatGPT integrations Via our public API and MCP integrations, your meeting intelligence now lives inside the tools you already use.

• Connect Fathom to Claude or ChatGPT

• Turn any meeting into a draft email, QBR, brief, or plan - no copy-pasting

• Available on all plans, Free through Business

🔍 Ask Fathom - now across your entire meeting history Ask Fathom used to answer questions about one meeting at a time. Now it answers questions about all of them.

• Ask "What are customers saying about our pricing?" or "What action items came out of my meetings last week?"

• Get AI-generated answers with citations linked to the exact transcript moment

• Your meeting history is now a searchable knowledge base

🖥️ Completely redesigned desktop experience

• Live summaries appear during your meetings, not just after

• Built-in scratchpad folds your personal notes directly into your post-meeting summary

• Less tab-switching. More focus.

📱 iOS app - coming very soon Not every important conversation happens on Zoom.

• Capture in-person conversations on the go

• Review meetings and access your full history anywhere

• Pre-download on the App Store shortly

Why this matters:

"The real value of meetings isn't the notes, it's the knowledge inside them." That's been our north star since day one.

We've always had two customers: the person in the meeting who wants to be fully present, and the leader who needs visibility across their team. This update serves both in genuinely new ways.

And all of this is available for free - because it's your call data, and you should make the most of it.

Drop a question below. We'll be in the comments all day. 🚀

- Rich & the Fathom team

Fathom, you beautiful beautiful people.

I started using Fathom two years ago and it completely transformed how I coach.

I'm a Career Coach with ADHD. Taking notes on client calls was always a struggle. Things got dropped. It pulled me out of the conversation. Fathom fixed that. I could stay fully present while my clients got notes, action items, and full call recordings they could actually revisit instead of relying on memory.

I also use AI to help clients quantify their career achievements. The transcriptions are everything for that. I can stay in the moment, stay engaged, and still capture every detail I need later.

But the gap was always timing.

I couldn't access any of that until after the call ended. So I ended up running Otter alongside Fathom just to get real-time transcription I could feed into AI mid-call. Two tools doing one job. Not ideal, but it was the only way to surface insights while the conversation was still happening.

And phone calls? In-person meetings? Those weren't part of the workflow at all.

Now all of that is solved.

Live transcription during the call. Phone and in-person conversations captured. Everything I was stitching together with workarounds, built right in.

Congrats team. This is exactly what I needed!

Question: It looks like the iPhone App is a separate paid app. Is there a way to access it as an annual Fathom Subscriber already?

I've tried a ton of similar products—we even temporarily switched to Zoom's "free" features—but we ended up coming right back to Fathom. It is hands down the best solution on the market. By far. With their new MCP server, it’s incredibly easy to integrate meeting content directly into our Customer Success workflows.

This is a great example of using AI to drive productivity, for real.
Kudos to@richard_white and the rest of the @Fathom team!

Hey Woww,

Fathom is a must-have tool, I’ve been using it for about a year now. I remember I was the only one using it at the company, and now all EAs and even some C-levels are using it.

Why? Because it records, summarizes, and the best part, you can ask it for exactly what you need from each meeting.

At the end of the day, I do a quick audit and review things like:

• What tasks were assigned to me?

• What’s important or what did I highlight?

I also pair it with another AI tool that helps me summarize my entire day in about 10 minutes 🙂

(FINALLY!) I have my OpenClaw connected to Fathom for transcripts, but have been frustrated at having to use Fathom + another note taker for non-zoom/Google meetings. Super happy I can bring this down to one tool that I like.

As someone who was a top 1% Gong user (per Gong), Fathom checks all the boxes and then some.

Well, most. Most of the boxes.
There are a few oddities (can't use the Zoom app without launching fathom is an example), but this is the easiest and lightest call recording tool I've used in a while.

Fun fact: we unsubscribed and re-subscribed once they opened their API.
Highly recommended.

I love Fathom. I have been using it for several years, and I have yet to find anything that compares. Today I went back to recording from 2024, went to "Ask Fathom" about a protocol I had done, and it was outlined in detail perfectly.

The customer service is fast and stays with you until you are clear.

I can't say enough about it.

As a life coach who does most of his work on Zoom, Fathom is absolutely indispensable! Not only does it record my sessions effortlessly, it provides me with huge value after the call in the form of transcripts, summaries, action lists (great for remembering homework I gave out!), and even the ability to ask Fathom questions for insights about individual sessions, all the sessions I've had with an individual client, or about patterns across the board, and so much more. Hard to imagine working without it!

I have been using Fathom for several years, and I've also left Fathom, trying to find a better solution. And I'm back! My favorite part is the Asana integration, which automatically pulls action items and sends them to Asana for me to follow up on.

We've tested a lot of call recording and note-taking tools over the years as a high-volume sales team, and nothing has come close to Fathom.

At this point, it's fully embedded into how our team operates. Every AE uses it on every call, and it has genuinely changed the way we run discovery, training, and follow-ups.

A few things that stand out for us:

First, the automatic summaries are incredibly accurate. After a call, we immediately get a clean breakdown of key points, action items, and next steps. This alone saves our team hours every week and keeps deals moving without anything slipping through the cracks.

Second, the ability to instantly pull clips is a game changer. Whether it's a strong customer quote, an objection, or a key moment in a discovery call, we can grab that snippet and share it internally in seconds. This has been huge for coaching, deal reviews, and aligning our team.

Third, search and transcripts are fast and reliable. We regularly go back into past calls to find specific details, confirm requirements, or prep for follow-ups. No more digging through notes or trying to remember what was said.

From a sales leadership perspective, it's also been invaluable for training. New team members ramp faster because they can watch real calls, see what “good” looks like, and learn from actual conversations instead of theory.

It also keeps our entire team aligned. Everyone has visibility into conversations, which makes collaboration between sales, solutions, and leadership much smoother.

Bottom line, this isn't just a recording tool for us. It's become a core part of our sales workflow.

We absolutely love it and highly recommend it to any team that runs a serious volume of calls and wants better visibility, better coaching, and better execution.

Pretty good. I'm surprised how many tidbits it can pickup. Gone are the days of having presenter/note-taker issue, as you know it's nearly impossible to do both WELL.

We have been using Fathom since the very beginning. We are champions for the product and mention it as much as we can. We have also tested a few other options, but always find ourselves coming back to Fathom. It has been reliable and saved our butts on soooooo many meeting details. Saving the link to the meetings in our project management platform allows us to kick off the meeting, and assign the project in no time. We love Fathom and promote it on our socials and podcast all the time. Thanks FATHOM!!

I use Fathom every single day and have no idea what I would do without it. It allows us to capture insights that inform everything from product and pricing to customer experience and context.

I've been waiting for these changes. So excited to check them out!

I've really enjoyed using it over the last couple of years, first with teammates at work, and then eventually found some use cases in my personal life. I love having a catalog of video recordings to look back on and track trends, conversations, little moments of joy with friends and colleagues alike. I'm excited about the MCP support that's coming out that will make it even easier to search using natural language. Excited to try out 3.0!

The bot-free approach is a huge deal. In finance, meeting context is everything — deal terms get discussed verbally, and if the notes are off, downstream decisions suffer. I host a podcast on financial modelling (ModeLoop Podcast on Spotify) and accurate transcription of technical finance discussions is genuinely hard. The Claude integration is interesting — curious whether it handles domain-specific terminology like DSCR, IRR waterfalls, and covenant structures well out of the box.

Regular user here. Love the product, transcription and UX is on point. Stoked to see them officially launch here.

Wow, would like to have a try in our next meeting. Congrats on the launch!

I was an early Fathom user: it's been keeping track of my meetings for years.

The fact that I can now pull every meeting I've ever had into Claude or ChatGPT and instantly execute on whatever I discussed in that meeting—with the whole history of everything that came before— is just insane.

What a world we live in.

Love, love, love Fathom - it saves so much of my brain power by recording all my meetings. I nevery have to look at my non-sensical scribble ever again! Great product.

I've been a Fathom customer for many years and these are the top features we've wanted. Awesome launch!

I've been using Fathom for three years and I have notes and transcripts from 500 meetings stored in the system. I've already been impressed with the account wide built in LLM chatbot and its ability to synthesize useful insights and summarize important points from across years worth of meetings.

I love Fathom!! I use it for all my meetings. The summary and action items sent out to everyone on the call keeps us all on task!

I love it so much! When Attio integration?

Long time user of Fathom here. there are several things about the new update that I find are a downgrade and can't turn off ...
1. A new window pops up when I start a meeting and give me notes that I don't need.
2. Another window now pops up after the meeting and gets in the way of my workflow.

I understand these can feel like a "value add" but they're really noise in my experience and degrade the experience overall. At minimum I'd love a way to turn them of.

This is the tool I consistently use the most. One of those things that makes you wonder how you survived before. Super stoked to try out the new features.

Fathom is the best!

I love Fathom, and I love this new update.

Amazing product, need to try the bot-free capture.

Everyone wants AI without seeing AI bots.

Do you have any upcoming updates planned for the API? I’m aware of the webhooks, but I’m also interested in whether there’s a way to pull meeting notes data directly.

Been a customer now for two+ years. It's an amazing product, and really excited about the new capabilities!

Claude Code Routines is a new automation layer inside Claude Code that lets developers configure scheduled or event-triggered AI coding tasks with no cron jobs, no servers, and no infrastructure to maintain.

The problem: Developers already using Claude Code to automate their dev cycle were still managing cron jobs, infrastructure, and MCP server tooling themselves. The AI was capable, but the plumbing was on you.

The solution: A routine is a Claude Code automation you configure once, including a prompt, repo, and connectors, and then run on a schedule, from an API call, or in response to an event. Routines run on Claude Code's web infrastructure, so nothing depends on your laptop being open.

Features worth noting:

Three trigger types in one surface.

• Scheduled routines — Triage, label, and post summaries automatically (hourly, nightly, or weekly)

• API routines — Each routine gets its own endpoint + auth token; connect to deploy hooks, alerts, and dashboards

• GitHub routines — Trigger on PRs via webhooks for reviews, triage, and auto-sync changes

Claude opens one session per PR and continues to feed updates from that PR to the session, so it can address follow-ups like comments and CI failures.

Who it's for: Engineering teams and solo devs who are already on Claude Code and have repetitive workflows they're currently handling with cron jobs, GitHub Actions, or manual scripts. Especially useful for on-call engineers who need faster alert triage and teams maintaining multi-language SDKs.

What repetitive dev workflow would you automate first if you had 15 routine runs a day?

P.S. I hunt the latest and greatest launches in tech, SaaS and AI, follow to be notified → @rohanrecommends

Seems like a good idea especially trying to compete with OpenClaw that runs continuously. Could be useful to run cron jobs to scan code databases to increase code quality especially with the volume of code that Claude outputs. If it says what it's going to do then I think its a good addition. Would be nice if it developed a report on what it's learned during those cron sessions and how it will implement those thoughts into deeper levels of model customisation

So Claude Code can now babysit your PRs overnight without your laptop staying open. For on-call engineers drowning in repetitive triage, that's not a small thing. Curious how the 15 daily routine limit holds up for bigger teams in practice. 🤔

I started using Claude Code for repetitive tasks and I'm saving 25 minutes per delivery. That's not a small number when you multiply it across a week. Now Routines is dropping and I'm actually excited — which doesn't happen often.

Love this! I use Claude Code daily to build my security scanner. Routines could be a game changer for repetitive scanning workflows.

Claude keeps showing up in more workflows lately, especially for coding and long-form thinking. Curious what you see as the biggest unlock with things like Code Routines?

Love that it runs on Anthropic's infrastructure so you're not babysitting your laptop. That detail alone makes it actually usable

I'm saving 25 minutes per delivery on repetitive tasks thanks to Claude Code. I can't wait to see what Routines can do.

Excited to hunt Intent by Augment Code today.

Intent is a developer workspace where agents coordinate and execute work end-to-end.

This isn’t a coding assistant. It’s an agent-driven development system.

Instead of prompting one agent at a time, you define a spec and a coordinator breaks it into tasks, delegating to specialists (implement, verify, debug, review) running in parallel.

This adds up to:
• Specs that stay alive as work progresses
• Built-in verification loops, not just code generation
• A full workspace (editor, terminal, git)

If you’ve been exploring agentic dev but didn’t want to build the orchestration layer yourself , this is definitely worth a look.

Running agents one at a time while keeping track of what each one did is exhausting as a solo dev. The coordinator layer is the thing that's been missing.

How well does it handle larger, messy codebases in practice, where context and dependencies aren’t always clear?

I've been using Augment Code on a large SaaS application with a strict domain-driven architecture, multi-tenancy, and dozens of interconnected domains. Most AI coding tools struggle with this kind

of complexity. Intent make it simple to launch coordinate agent to develop new feature in parallel!

The code editor is really a great concept because I do have to maintain a separate workspace just for the nodes. And I hope it has voice-to-text built-in.

This sounds cool, tell me, how I can trust on quality of this agent for work?

Wrote up a post on how our teams collaborate within Intent. We've been able to effectively eliminate the designer/developer handoff. More details on the process, screenshots, etc: https://lukew.com/ff/entry.asp?2148

the isolated workspace approach is smart. removes the "works on my machine" problem entirely when agents are doing the building. how do you handle dependencies that need specific system configs or external APIs during the build process?

I like this. It seems very interesting. CLI code understanding and agent-driven development make a lot of sense. I'm just wondering, does it do any browser work? That's where most of my time currently gets sucked going back to the browser and seeing if everything has been implemented okay. I think that would be a great problem to tackle anyways. Love the product, and we'll give it a try. Best of luck.

So does this work with multiple repos? Like old legacy code or needs a monorepo to work well?

Really interesting direction. Moving from copilots to coordinated agents working from a spec feels like a big shift. How do you see teams defining good specs so the output stays reliable?

Congrats on the launch! Wondering what’s its integration capabilities with many common SDLC software, because building in isolation is great until you need to do some real work.

This is a complex thing. Who is the main target audience?

@byalexai the parallel execution model is interesting — when the coordinator breaks a spec into tasks, how does it handle requirements that turn out to be underspecified only after an agent starts working? Revert and re-plan, or does it try to resolve in-context?

Lovable Desktop is a fast, lightweight app designed to help you organize projects with tabs and connect directly to local MCPs, solving the friction of scattered workflows and limited local integrations.

What makes it stand out is its local-first approach with MCP support, combined with a clean tab-based workspace and native keyboard shortcuts for speed.

Key features:

• Project tabs for multitasking

• Direct local MCP connections

• Native Mac performance & shortcuts

Perfect for builders, developers, and teams who want faster, more streamlined workflows on desktop.

If you’re building across multiple projects and want tighter local control, this is worth checking out.

P.S. I hunt the latest and greatest launches in tech, SaaS and AI, follow to be notified → @rohanrecommends

Desktop apps for AI builders felt inevitable - browser tabs don't cut it when you're in deep flow. The MCP support is what makes this genuinely powerful. As someone building security tools for vibe-coded apps, I'm curious: does the desktop version add any security sandboxing for the generated code preview?

The local MCP connection is the feature that changes everything, being able to connect your own tools directly without going through a browser tab is a huge workflow upgrade.

As an indie iOS maker who lives in Xcode, I'm curious: is there any plan for Xcode or Swift-specific integrations, or is Lovable primarily focused on web app development for now?

I'd love to try it out, but ... safari can't open the page http://127...

Gemini Robotics-ER 1.6 is the reasoning layer that lets robots like Boston Dynamics' Spot read analog gauges, count objects, and confirm when a task is actually done. Available now via the Gemini API.

I'm hunting this because there's a gap between "robot that follows instructions" and "robot that reasons about what it sees". That gap is exactly where industrial automation keeps getting stuck. ER 1.6 directly bridges that gap.

The problem: Most robot AI can execute. Very few can verify. Knowing when a task succeeded, reading a pressure dial in a poorly lit facility, or identifying the correct object among 40 similar ones requires embodied reasoning, not just vision.

The solution: A vision-language model that handles pointing, spatial counting, multi-view success detection, and instrument reading as first-class capabilities. It can call tools natively and chain reasoning steps to solve complex physical tasks.

Key capabilities:

• Spatial pointing: detect objects, map paths, find grasp points

• Success detection: confirm tasks across multiple camera views

• Instrument reading: read gauges, sight glasses, digital displays (93% accuracy)

• Agentic tools: integrate Google Search, VLA models, custom functions

• Safety constraints: respects material and weight limits

Who it's for: Robotics engineers, hardware AI teams, and developers building autonomous inspection or manipulation systems. Especially useful if you are integrating AI reasoning into industrial or field robotics.

P.S. I hunt the latest and greatest launches in tech, SaaS and AI, follow to be notified → @rohanrecommends

The spatial reasoning piece is what makes this interesting. That's been the hard problem for physical AI for a long time

When you have multiple Claude sessions running at a time, do we get multiple pages, or does the page screen get increased with each Claude session?

Having a beeper UI on my desktop in 2026 was not on my bingo card for the year.

This is super cool.

One suggestion: you could consider color-coding the statuses based on whether there's input needed from the user. (eg. "done" -> green, "error" -> red, "needs input" -> yellow).

This looks really great. I hope we have something similar for Windows Laptop too, sooner. Just in case, if there are multiple Claude sessions going on, how it will react, I am curious to know that.

Seriously, the state that attracted me the most was the error: “Something went wrong.”

A lot of times, I only see a summary at the very end of the run, and I end up getting lost in what actually happened during the execution flow. The real issue is that when there are multiple errors and the claude tries to handle them on its own, it becomes painful on my side. It affects the whole workflow.

Instead of finding out only at the end, I’d rather get real-time signals about errors as they happen, so I can step in immediately and help, rather than reacting after everything is already done. COOL!

Looks very cool! Will give it a try!

This fills a real gap - I've caught myself staring at the terminal waiting for Claude to finish way too many times. The four auto-accept modes are the right call, YOLO mode for scaffolding and Strict for anything touching prod. One question: does the LCD status show which tool Claude is currently using (file read vs bash vs edit)? That granularity would be genuinely useful for knowing when to intervene.

Love the retro pager concept. It makes something pretty technical feel way more approachable. How did you decide what states and signals were most important to show in real time?

As a Claude Code user- this is clever. Does it work with the new agent/skill features too?

naming the top trust level "yolo" is the most honest ux decision i've seen in a dev tool. been running claude code for months — you always end up at two extremes: approve everything manually or just run

  --dangerouslySkipPermissions and hope. strict to yolo as an actual dial makes trust a real decision.

This is so fun. Did you ever think about trying to connect this with a real beeper? I also can't believe this is your first open source project. I'm definitely looking forward to seeing what you're building. This was great. Well done @vecartier!

This is really a nostalgic concept. I remember having a disposable camera when I was really young. Having that same concept in the mobile itself feels really awesome. I hope you also launch one for the Android phones too.

Really love this concept. Turning photos back into something you wait for instead of instantly reviewing feels surprisingly powerful. Have you seen people change how often they take photos because of the limit?

Love this concept for so many reasons. Number 1 is my 2 year old daughter. Every time we take a photo of her, she has to see it immediately. This is a learned behavior from watching her mom and dad. Always taking several photos to get the perfect one instead of living in the moment, and whatever photo we snapped we got! Just one question, why did you land on 12 photos?

Adore this concept, and design is outstanding ::chef's_kiss::

With so many products designed to make photos look like film, I love that Roll makes photography feel like film. Aspect ratio is a cool feature: I can make my images look like 120 film! Love the tips. How do I turn on gridlines (tip 10)?

I'm also curious (like others, below) about what behavior patterns you see as DAUs grow: are users tending to shoot all 12 images at once, or being more intentional and thrifty with their "Rolls".

Good luck! Will be following enthusiastically!

Back in the day, where disposable cameras used to make the things feel alive because they were limited, and you could only take so many shots. With digital cameras nowadays, the amount of pictures you can click is unlimited. I love that you are bringing something new, and yeah, congratulations on the launch!

this hits different than all the AI photo apps flooding the market. reminds me of waiting to get film back from CVS and forgetting what you even shot. the anticipation was half the experience. curious if you're tracking any wellbeing metrics around reduced photo anxiety or phone usage patterns?

Hello, Claude. I really like the idea of this app - I remember the old days, too! I was wondering if you have to take all 12 photos in one session, or if (for example) you could take six photos today and six tomorrow? Good luck with your launch. Thank you!

This is awesome, I had an idea like this a few years back, but could never figure out the one filter "fits all scenarios" implementation. I think it's cool that you didn't try to turn it into an entire social media app. It's nice to just have a tool that serves its one purpose. Love how minimal the UI is as well, great work!

Love how this leans into the constraint instead of fighting it. The "limited shots" friction is the whole reason disposables felt magical in the first place, you actually thought about each frame. Curious though, do people wait to "develop" the roll or does it break the spell when they peek early? And is there any social layer where friends see each other's rolls when they drop?

Are you looking at anything around sharing "cameras" with contacts? If you think about weddings, where there used to be a trend of leaving a disposable camera on each table for people to use and get candid shots of the event, the bride and groom got the surprise doubly by not knowing who took the pictures or what they took them of. Having something similar, where the coordinating users could send invites to contacts, contacts would interact on their phones, and all the images taken would return to the coordinator for review/curation for a shared album the contacts have access to. I could see something like that really amping up the fun of the event use of the app and conceivably allow more gamification (ratings, reviews, comments in album) and possibly making the app more viral.

Reminder, if you've played with the app today, would love to get your feedback via this form! Thanks so much!
https://docs.google.com/forms/d/e/1FAIpQLSehEIMAmh51PyhdyYTpVK3-0CV0fr1osEtMkpNJ2W6LHY7W2A/viewform?usp=header

This is actually insane. I'm curious to see what drove you to build this. It's something that I always hoped that I had but never really thought was possible. Gonna test it out

Cool mate! Is it available everywhere? Anyway I feel it's a great deal for everybody here cause it helps founders to better perform in a daily basis by giving more and more speed. Wish you all the best here!

This sounds pretty cool. I thought you could do this already with certain technologies, but I've never been able to do it. So if this makes it easier for that, that's brilliant. My Ethernet maxes out at a gigabit, and my internet connection is 8 gigabits. If I could have a bit of Wi-Fi on top, that will go a bit quicker still, and that's already a big win.

Please correct me if I'm wrong but this feels like multipath TCP but at the application layer. If so the why browsers don’t already do this?

The upgraded Google app for desktop is now available for Windows users globally in English.

Use a simple keyboard shortcut (Alt + Space) to instantly find what you need—information from the web, your computer files, installed apps, and Google Drive files—all from the Search box. With AI Mode in Search and Lens built right in, you can ask whatever’s on your mind or search what’s on your screen to get helpful AI-powered responses with links to the web.

I was so disappointed with this.
I am on their PRO plan and this desktop app looks so basic and useless.
Like it's a chat box hooked to a server where the intelligence is - Not even comparable to Claude Desktop.

I'm also very disappointed in Gemini Live mode - so compressed responsed.
You can't discuss with this, it feels like it doesn't want to discuss with you and doesn't do any useful research.

The only good AI product of Google that I know is Veo 3.1

Google lens whatever is on my screen without pointing my phone on it like a boomer thats cool fr

Congrats on the launch. I just started checking for the same and found this out. Nice, pretty good idea. Do you also auto-publish to LinkedIn, or should I download and publish manually?

You know, I tested it, and it did something pretty good. First time, it wasn't amazing. It made a lot of assumptions, which I didn't want it to. I prompted it, and now it came up with a decent pitch. The colours weren't quite right, 'cause I think I gave it the wrong URL for the best branding colours, so I injected some brand colours, and after a third iteration, it was great, and it's now live here -> https://www.linkedin.com/posts/fruey_a-daily-zen-ritual-for-your-photos-ugcPost-7450140778166882305-oL1f

It could be useful. We'll definitely try it out! Beyond LinkedIn, do you see this expanding to Instagram carousels, Twitter threads, or newsletters?

Pulling brand colors and fonts directly from a URL is the feature that makes this actually usable. Every other step is already easy. That one isn't. Congrats on the launch.

How close are the generated carousels to something people can publish straight away without editing?

Steali...i mean automatically extracting brand kit just by URL thats crazy feature for us lazy marketers, love it🔥

Congrats on launching! LinkedIn carousels are such a time sink when you're a one-person marketing team trying to stay consistent. The Brand Kit auto-import alone would save me so much back-and-forth. Can you customize the number of slides or does the AI decide the format based on the prompt?

Congrats on the launch.

I however have a small concern. At what point would using your tool start to hurt differentiation? Like when everyone has access to similar AI-generated carousel structures, how will my company stand out from 1000s of others using the same layouts on repeat. Without a creative direction it might all just blend into one

Hey, cool site. Just FYI: I noticed there is a French text in the middle of English content:

Hi Product Hunt! 👋

Cansu here, founding designer of Defter Notes.

Launching version 2.0. on Product Hunt has been on my list for a while, and it feels right to do it now that it has had a few months to breathe and grow.

Defter Notes began with a simple frustration: most note-taking apps ask you to think like the app, instead of the other way around. I and my partner Caner wanted to build something that felt more like a messy desk with sketchbooks rather than a database. That idea is still at the heart of everything we do.

2.0 was a big rebuild as we tackled some performance uprades, had a chance to redesign UI elements and implemented new features like iCloud sync for the whole Apple ecosystem. We have been refining it with help from our beta testers and shipped it quietly in November.

A few things that have meant a lot to us along the way: Last month Defter Notes was featured as App of the Day in the US, Australia, and New Zealand, and was picked as one of Apple's 'Apps We Love.' It was a recognition we have been looking forward to since the first version and we were happy to see it arrive very early with version 2.0.

So, to celebrate our Product Hunt launch, we’re offering 1,000 promo codes for a discounted Lifetime Access plan. It's available during launch week until April 19. Follow this link to claim the offer: https://apps.apple.com/redeem?ctx=offercodes&id=1570783518&code=DNPH

Thank you for supporting our work! And if you try it, I'd really love to hear what you think. 🙏

If you have questions about the app, or would like to chat, feel free to join our Discord community.

Cansu

The spatial handwriting combo is interesting. Feels like it gets at something typed notes can't quite replicate, my best thinking still happens on paper and i've never figured out why typed notes go dead on me halfway down the page. Curious how you landed on spatial as the organising principle, was it from watching how people actually use notebooks or something else entirely? Good luck with the launch.

LinkedIn carousels are huge right now. Does it pull brand assets automatically or do you feed them in manually?

how does astra treat this raw sensitive data (which it processes)?

This is a critical problem in financial services. When we build project finance models for renewable energy deals, the data flowing through them — tariff structures, counterparty financials, tax equity terms — is highly sensitive. The idea that AI agents can reason on tokenized data without ever seeing the raw values is exactly what regulated industries need to adopt AI safely. I publish financial model templates on Eloquens (https://www.eloquens.com/channel/samir-asadov-cfa) and data privacy in model distribution is always a concern. Would love to see this applied to financial modeling workflows.

Yo @obed_mpaka1 @Astra quick question.

I’m a researcher for H1Gallery newsletters (you can google us).

We’re featuring Astra in the April 17 H1Gallery issue. H1Gallery highlights excellent homepage headlines, and “AI agents handle sensitive data every day. They shouldn't see it.” really stood out to us. Its clear and compelling.

I wanted to reach out to see if you’d be open to sharing a quick comment on the copywriting strategy behind that headline and the broader messaging. We’d love to include a short note from your team on how you approached it.

Totally optional, of course . The feature is happening regardless either way. Our readers love to hear from the creators behind the headlines tho. And sorry for the late notice!

Thank you so much. Love the product!

@obed_mpaka1 the tokenization-before-prompt approach is interesting — what happens when the agent's reasoning output references a token and you need to log or audit that decision? Does the audit trail show the real value or does it stay tokenized end-to-end?

The 'powered by your private data' part is what matters here. Most agent platforms force you to feed everything into someone else's cloud. How do you handle data residency — can everything stay on-prem, or is there a hybrid option for teams that need both?

One of the coolest launch of the day! Btw once it identifies bottlenecks, how are fixes applied like automatically, suggested or human in the loop???

@renchu_song the 1M token burn on a single LLM call with no visibility is a very relatable war story — does ClawTrace surface cost attribution per agent or per task/subtask? Trying to understand if the granularity is enough to catch runaway sub-agents before they crater a budget.

Cool project but you guys def need a design work for your branding/logo. aye human one

Hey PH 👋

Max from Playbook Team here.

We've spent five years building the best storage for creatives - and watched our users spend hours doing things that should take seconds. Finding that one hero shot. Reorganizing a campaign folder. Prepping a client share link. Every. Single. Time.

Playbook Intelligence is our answer to that. Not just smarter search - a conversational layer built into your files. You ask, it acts.

We're launching with three things today:

• Talk to your files (organize, share, bulk edit)

• AI Organize: propose and apply structure automatically

• Sidebar Search: find and act in one motion

We already support 2.5M teams and individuals including folks from Warner Bros and The Vault Stock. This is the feature they've been asking for.

Now tell me - what would you ask your files first?

Oooh! 😍 — also thinking this will feel much better to run from the Playbook interface, rather than giving an agent unlimited access to my whole computer to organize files

Here's how you learn something. You practice and you get it wrong. Someone tells you why. You try again & you keep going until it sticks.

But that's not what happens at school. There aren't enough exercises to go around, and even when there are, there's no guarantee anyone's checking. A teacher with 25 students and 45 minutes is doing triage, not tutoring.

Practice without feedback isn't practice, and the system doesn't have nearly enough of either.

Lexie closes that gap. Take a photo of your biology chapter on the circulatory system and Lexie figures out what an exam would test from it. Then it gives you different ways to practice until you know every valve and vessel. Then spaced repetition schedules when to review it all. So you get everything learning science knows about effective practice without having to know any of it. No surprises on exam day.​​​​​​​​​​​​​​​​

Most edtech gives students gamification theater: points, streaks, engagement mechanics that feel good but teach nothing. I stripped that out. Photo determines content, AI determines difficulty, testing reveals gaps. No escape routes. Learning isn't consumption. It's construction. And construction is messy, effortful work.

I designed it so a 10 year old can go from app store to first study set in 30 seconds. No account, no login, no onboarding. All the friction is in the learning, where it belongs. Also, no ads, no trackers, no data selling. Photos stay on device. When you're 16 and need help with geology, you shouldn't have to trade your data for it. Making money from subscriptions, not from students.

Spaced learning is a lost feature with these study apps, devs seem to be forgetting the forgetting curve (ha) identified by Ebbinghaus. ALso the concept of protected data is huge, as is the positioning of simplicity, so that our kids could use it. Love the authenticity Elina.
Will be following this one.

Removing streaks and points is a deliberate choice — most apps lean hard on them. Do you see adult learners using this, or is it really built for students preparing for exams?

Been following your journey on LinkedIn. My daughter might soon need something like this (she's 4) and when that time comes, as a parent, I know I'll try Lexie because I know your heart's in the right place making this!

Hey, Product Hunt Community!

Meet Pane Studio — Beautiful Screen Recordings, Minus the Editing Headache 🎥

I built Pane because Windows deserves a screen recorder that feels premium and native — fast, fluid, and beautiful out of the box. No clunky interfaces, no laggy previews, no compromises. Just the smoothest recording and editing experience on Windows, built for people who want their demos to look as good as the product they're showing off.

I believe screen recordings shouldn't look like screen recordings. They should feel polished, cinematic, and on-brand — without you ever touching After Effects. That's why Pane lets you restyle your recording after the fact — resize the cursor, smooth the motion, reframe the shot. No re-recording needed.

What's inside Pane Studio 🎬

🎥 Effortless Recording – Full screen, specific area/window, or app. Multi-monitor support. Mic + system audio baked in.

✂️ Built-in Editor – Cut, trim, speed up, crop, and switch aspect ratios (landscape, portrait, square) without leaving the app.

🖼 Beautiful Backgrounds & Padding – Pick from curated wallpapers or drop in your own. Add shadows and padding for that polished, "designed" look.

🔍 Smart Zoom Effects – Auto-zoom on cursor, or set zooms manually in the editor. Your viewers' eyes go exactly where they should.

🖱 Cursor Magic – Resize the cursor after recording, smooth out shaky movement, hide idle cursors automatically, loop cursor position for seamless demos, and apply custom styles. No re-recording because your mouse wobbled.

🎭 Cam Layouts & Masking – Place your webcam in customizable layouts with clean masking options to hide sensitive info or highlight important parts of your screen recording.

📐 Export Presets & Single Frame Exports – Up to 4K 60fps. Ready for YouTube, Shorts, TikTok, LinkedIn, or handoff to a bigger editor.

🔒 100% Local – Your recordings never touch a server. Everything stays on your machine.

💡 What's Coming Next

This launch marks only the beginning. In the coming months, we’ll be focusing on:

✅ Shortcuts Capture – Quickly capture actions using customizable shortcuts for a smoother workflow.
✅ More cursor styles, transitions, and motion blur – Additional visual options to make recordings feel smoother and more polished.
✅ Transcript generation – Automatically generate transcripts from your recordings for easier review, sharing, or subtitles.

Pane is the tool I always wished existed — and I'm so excited to finally put it in your hands.

I'd love your feedback. Break it, stress-test it, tell me what's missing. I'm here in the comments all day. 🙌

This is the exact stuff i keep putting off for my own launch coming up. The polished demo video problem is real, Loom recordings look amateur but a full video shoot is a different job entirely. Does Pane handle the in-between, clean narration, clean cursor movement, simple cuts? Or is it more polished screen capture with effects on top? And is there a template library for common launch video formats?

Really interesting approach, especially keeping everything local. That’s a big plus. Do you plan to add collaboration or sharing features down the line?

Demo videos are a pain to make. What makes this different from Loom or Arcade?

Been looking for a Screen Studio equivalent on Windows for a long time, glad this exists. Genuine question: there are solid open source alternatives that handle recording/editing for free. What's the case for $10/month: is it the cursor smoothing and auto-zoom, or is there something in the workflow that genuinely saves enough time to justify it for a solo builder?

This looks genuinely useful. Most screen recorders are fine at capturing, but the annoying part is making the recording look good afterward. Pane seems to actually care about that part.

Also, I like that it’s fully local!
That’s a big plus for anyone recording sensitive stuff.

Curious how it holds up on longer recordings.

Congrats on the launch! 🚀

Hey Product Hunt 👋

Introducing: Doz - a medication reminder that actually works in real life.

I ran into a small but frustrating problem.

Taking multiple medications daily — some before meals, some after, across different prescriptions — quickly became confusing.

I remember moments where I just stared at my reminders, thinking:

• “Was this before or after food?”

• “Which prescription does this belong to?”

• “Did I already take this… or not?”

I tried alarms, notes, and a few apps. But they all treated medications like generic tasks — not something structured around how prescriptions actually work.

So I built Doz — first for myself. Over time, I refined it based on real usage, and decided to share it.
Instead of forcing myself to remember, I wanted the system to match how medications are prescribed and taken in real life.

What makes Doz different:

• 📋 Group medications by prescription — less confusion

• 🍽️ Tied to meals → no manual time calculation

• 🚨 Reliable reminders (even in silent mode)

• 🔐 Privacy-first — no account, data stays on your device

Who Doz is for:

• People managing multiple medications

• Anyone who struggles to keep track of prescriptions

• Those who need reliable reminders they can trust

Free to download, with optional Pro.

🎁 PH users: use code DOZGIFT for Doz Pro Lifetime (limited slots)

Would love your feedback — happy to answer any questions 🙌

👉 getdoz.app

Anything built from a personally suffered pain point is going to be the best product, really love the app Huy. Before becoming a solo dev myself, I was a PT in the Long term care setting. We often discharged patients home alone, and in some situations a tool like this, could have been a literal life saver. On that note, I hope the UI/text has the option of being "Jitterbug" big haha.

Fellow solo dev here. The prescription-first approach is the right call. Most reminder apps treat every pill the same regardless of context. The meal timing sync is the detail that shows you actually talked to users.

I hope we get access to all the models with this flat rate, with some sort of limitations from the tokens, or you are not doing that and giving a flat rate on all kinds of models.

I hope we get access to all the models with this flat rate, with some sort of limitations from the tokens, or you are not doing that and giving a flat rate on all kinds of models.

what it means for $1.00 each in your pricing?

Awesome that we can create tabletop miniatures but can do we have dynamic poses on figures?

Nice, love it. From what team size do expect this to become valuable? We are currrently with 2, so dont see the use case right now.

I co-founded Collabute after watching brilliant PMs spend half their week explaining decisions they'd already made to developers, to stakeholders, to themselves. That's not a productivity problem. That's a memory problem. We solved it.
We would appreciate all the support