... your votes, and the new version goes straight back on the board. Last time over a hundred requests were shipped before midnight, and more than 8,000 points went out with them. Supabase and TypeSafe AI, makers of Jev, are both back with credits. Not launching? Come anyway and file requests. The whole thing only works if people are trying the products and saying what's missing, and on the day your feedback is somebody's next feature. Schedule your ... ... from Olia Nemirovski(@olia_nemirovski): Count us in for Friday with Tobira Site Agent! One thing I haven t seen asked yet: are points flat per shipped request, or weighted by the size of the work (a quick fix vs

... Jev is only as good as it is calibrated to your use case. We are launching Brussle to make it super easy to calibrate and as part of the launch we are super excited to give 4K free jev

... Product Hunt! Michael here from ZeroClick. We've added Jev-powered Seller Agents to our Agent Storefronts to help AI agents buy from your business faster, with fewer tokens spent figuring out how. A buyer agent shares its goal, budget and available payment methods. Your seller agent turns that context into concrete purchasing instructions, so the buyer spends less time searching documentation and making trial-and-error requests. In our testing, buyer agents used up to 27% fewer tokens ... ... options or completing the purchase? We'd love to hear a specific example! Comment from Palmer McCutcheon(@palmm): The team had a lot of fun building this one! Cool to get to the edge of what's possible with Jev

... Rikin, maker of dotpals (a small open-source pal that shows what your AI coding agents did). Most people seem to use Jev for lead-gen and booking calls. I'm using it for something different: as a judge for test results. My coding agent says "All tests pass. Done!" far too often. dotpals reads the test runner's own output first, and when the output and the exit code disagree, it sends Jev

... rule: memory has to be allowed to say no. What changed: - Recall returns supported, related, or "That isn't in SuperLocalMemory." - An answer check asks whether the memories found actually answer the question: Laya on your Mac, or Jev online with your consent. - One memory shared by Claude Code, Codex, Cursor and any MCP agent, and your agents can message each other. - Pictures and PDFs you find by describing them, with Google's EmbeddingGemma 2 running locally. - Roles, an audit ...

... platform. Someone shares a pitch as a Google Doc, so the email comes from Google. Or they sign you up for their Zoom webinar, so the invite comes from Zoom. You can't send Google or Zoom to spam. Jev is what finally fixed it. It doesn't write anything, it just tells you how likely each yes/no answer is, and you decide where the line goes. Per email I ask five yes/no questions in one call: did a person ... ... model. On October 2 there were 92 emails in the inbox. After the sort, 2 were left: a Google Workspace invoice and one reply I actually needed to answer. It runs every five minutes now. What are you using Jev

... look after, so the devops side is close to nothing. Coding stays on Claude Code, but on a flat monthly plan, so that cost doesn't grow with usage either. For a few features we've also started using Jev lately. And I watch API usage all the time, with a phone alert when it crosses a threshold. Comment from Ilian Madzharov(@ilian_madzharov): It's an interesting challenge. I was recently reading an article (highly recommend this blog ...

... Today is all about shipping velocity. You can leave requests for the products launching today and they'll build and launch the features in real time! Appropriately, given the speed theme, we've partnered with TypeSafe AI, makers of @Jev , a frontier model at real-time speed, and @Supabase . Each participant will need real, actionable feedback Rules Only featured launches are allowed to participate A launch s point score score = normal score + shipping points Each shipping request ... ... user can complete requests up until 10pm PT on the day of the contest. All requests must be verified by Product Hunt to be awarded points Winners will be chosen from the top launches. Prizes will include Supabase + Jev

... months, submit your launch by midnight Sep 30. Schedule it here: https://www.producthunt.com/post... Be prepared to build and ship multiple features in real-time. Appropriately, given the speed theme, we've partnered with TypeSafe AI, makers of @Jev , a frontier model at real-time speed, and @Supabase . Prizes will include Jev

... inspectable and interactive. Try it out by creating a local AI app today with https://www.producthunt.com/prod... And share with us what total transparency AI feels like! Concurrent agents and live model state during inference: Entailment based context admission (JEV style decision on what to read from a fetched web-page): Comment from Matheus Paranhos(@matheus_paranhos1): I get where you come from but shared working memory also makes the review trail harder. I ve had parallel agents reach different ...

... tokens, and new accounts get free trial credit. The playground needs no signup: https://meragpt.com/playground On the open typed-decisions benchmark on Hugging Face (zero-shot table), it scores 0.768 accuracy, ahead of Liquid d1 at 0.742 and Jev at 0.727. We maintain that benchmark. As a small demo on a synthetic folder, it sorted 184 desktop files into 8 folders from file names alone, 96.2% correct. What decisions in your product would you route to something like this ...

... them. For a post, those who stopped also say what they would comment. The page splits the reasons into ones about the text and ones about the audience, so you can tell a weak opening from the wrong readers. Jev also answers a few yes or no questions about the text itself: Is there a concrete number, name or example? Is it clear what the reader should do? For a listing, does the first sentence say what is for sale ... ... work in IT and are into startups" is 238 people, "people over 60" is 1,398. The town has no data on cities or company stages, so words about those do not count. The Audience tab shows how Jev

... TypeSafe launched Jev last week. It is a System One model meant for typed decisions inside software (classify, route, score, extract, branch), not chat. I am not on the team. I am trying to figure out if this actually changes how products get built, or if it is just another structured output wrapper with better marketing. The pitch in plain terms: you give it unstructured state, you define the output schema up front, you get calibrated probabilities back in tens ...

... vibe coder the docs are really helpful. Comment from Marriane Akeyo(@marriane_akeyo): The mesh protocol especially where agents get to communicate with each other directly, combined with the fact the framework is model agnostic and even works with Jev is what I really like about Jarviscore. Comment from Elizabeth Nyambere(@elizabeth_nyambere): Loving the cdp attach for browser automation. watching the agent take over when needed makes it way easier to trust ...

... time to sit wondering. Your guess makes sense to me, a product forum has a much smaller team behind it than p/general. All of mine went to p/general or p/vibecoding, so I've never seen that side. Did the Jev one come back with any note, or just the approval? Comment from Hisashi Space(@hisashispace): @abdullah_javaid3  that fits what happened to me after too. The two I got through since both had a hard number right ... ... anyone else's, so I haven't really tested it. Comment from Hisashi Space(@hisashispace): @abdullah_javaid3  Yeah, once. In general mine come back within a few minutes, 20 at most. But one I posted in the Jev

... really do need an LLM to judge something, I get other LLMs to check it. Like Claude makes the first call and Gemini double checks. Sometimes I throw Codex and Grok in too for a triple check lol. Jev just came out and looks good too. Comment from Magnus Ditlev S rensen(@magnus_ditlev_sorensen): @alex_goldwyn  Yeah, that makes sense. If someone needs to verify the result every time, it often ends up taking more work than doing ...