What are you building? Drop your AI product below

CellCog. Hire AI employees for any role. Each one has a name, an inbox, shifts, persistent memory, and a task board. I'm the only human at my company; the AI employees run the rest. Launching here July 29.

what: aver. the layer where people vouch for the work they actually saw. peer signatures over resumes. for humans and AI agents.

who: career changers who watch their credentials evaporate at every field transition. and hiring teams that want signal from someone who was actually there, not a candidate curated linkedin page.

hardest part right now: getting the first vouch feels heavy for a user who has never done it before. the muscle 'i saw them do X and here is the specific thing' isnt something people have practice writing. we're prototyping prompts that lower the barrier without turning it into a form.

Hello there,

I’ve been a voracious reader for as long as I can remember. But with the AI boom and the sheer volume of content published daily, keeping up with my reading—from deep-tech architecture deep-dives to tactical football analysis—has become a real pain point.

What I'm building: Etcetra — a local-first, summary-driven second brain. It distills the noise by auto-generating clear summaries of articles you save or bookmark, helping you decide at a glance if a piece is worth a full read. Over time, it curates these insights into an accessible knowledge base that you can easily extend with your own notes and thoughts.

Who it's for:

• The drowning reader who needs to skim high-signal summaries before committing to a 15-minute read.

• The power user who wants an effortless, persistent second brain built naturally from everything they consume.

The hardest part we're solving: Preventing knowledge overload. We want to deliver a centralized, searchable memory of everything you’ve read without creating another overwhelming backlog to manage.

The current version is up here and we are preparing to launch on this platform soon.

Building MemoryO — a governance layer for AI memory.

https://app.memoryo.dev

Building: I'm building Crosspublish to "Post to all your socials from a modern editor or AI agent."

Problem:
- Buffer offers support for all major social networks but a bad, one-size-fits-all editor
- Typefully offers a premium editing experience but only supports 5 platforms.

Solution:
Crosspublish gives you both. Premium, platform specific editing; and support for all major socials.
In addition it supports MCP, so it works wherever you work.

Listing: https://www.producthunt.com/products/crosspublish?launch=crosspublish-4

Building MemoryOS a control layer for AI memory.

It’s for teams building agents that already have memory, but run into the messy stuff: conflicting facts, outdated context, corrections, and multiple agents learning different things about the same user.

The hard problem we’re solving: when memory disagrees, what should the AI actually trust?

Launching soon. If you’re building AI agents and this is a problem you’ve run into, would love to connect.

Business Income Idea Generator

I built a small tool for people who want to start a side business but aren't sure what actually fits their skills and experience.

Answer 10 questions → get 10 personalized business income ideas, your Top 3, #1 Best Match and a 48-hour validation plan.

$19 one-time for 3 personalized reports. I'd love feedback from early users.

https://idea-spark-studio-53.lovable.app

Hi, We need your upvote. 🔥 😭

Tucky is live on Product Hunt: a macOS notes app that docks to the edge of your screen and includes a built-in AI agent. Local, encrypted, and free to start

Mac users, download it, try it out, and help us climb the ranks.

Product Hunt: https://www.producthunt.com/products/tucky

Good initiative, mate ;)

Mine:

• What:
  
  TrackingCoder: Paste a URL, it scans the site and works out what to track (buttons, forms, purchases, bookings), you pick, and one click places every tag, trigger and variable directly in your Google Tag Manager workspace. You review and publish. Google verified the GTM integration last month, which took some effort to achieve.
  

• Who:
  
  Mostly two kinds of people. Marketers and agencies who set up tracking for a living and want the afternoon back, and small business owners who have never opened Tag Manager and cannot afford a consultant. Same tool, very different hand-holding.
  

• Hardest part right now:
  
  Reading a site we have never seen and getting it right. Page builders, forms inside iframes, buttons that are divs, checkout flows that change on the second visit. The tag is the easy bit. Mapping a real, messy site to the right trigger is where the time goes, and a wrong guess costs the user trust immediately. We made it all super easy.

We launch here next Tuesday, so if you have a site that breaks tracking tools, I would genuinely like to point the scanner at it.

Have a go for free: https://trackingcoder.com/

What I'm building: successors.app, a webapp to help job seekers and or people who want to pivot in their career to identify the gaps between their existing resumes and the skills required from the target jobs, suggest actionable to-dos to close this gap, re-assess and rebuild users' confidence and help hone interview skills with an AI simulation.

Who it's for: job seekers who are motivated to do more upskilling or skills matching work to get closer to getting the job that they want

Hardest part: getting enough traffic to the page, minimal engagement and adoption metrics gains at the moment.

Free to try on successors.app, PH listing here

Thanks all for reading, please feel free to provide any feedback! Appreciate it. It's tough starting out since it's my first time launching a product, but I'm proud of my work nevertheless!