I woke up to a nasty bill because an agent looped overnight. All the usual tools (alerts, soft budgets) only reacted after the damage was done. So I built baar-core: lightweight library that checks budget before every API call.
Raises exception locally
No DNS, no TCP $0 charged on overrun
Semantic routing (84% savings)
Per-user SQLite quotas Just launched on Product Hunt: https://www.producthunt.com/post... Open source: pip install baar-core Would love feedback from anyone running agents or multi-tenant LLM apps in production.
I built NogginLogger because every calorie tracker I tried died the same death scanning barcodes, scrolling through 40 versions of "grilled chicken breast," and giving up by Tuesday. The friction wasn't the math, it was the input. So I made one where you just say what happened: a five-second voice note gets transcribed by Whisper, parsed by Claude into a structured entry (category, calories, macros, tags), and the dashboard shows calories consumed vs. burned in real time. Exercise calories are calculated server-side from your body weight and MET formulas, not AI guesswork. Once I had the voice-to-data pipeline working I started logging everything spending, mood, sleep, weight, even golf scores all in one place. What surprised me wasn't the tool, it was the accountability: when you can't hide from the data, you make slightly better choices, and slightly better choices compound. I lost 20 pounds in the timeframe I'd planned, with no dramatic diet just say it now, see it later. Set your goal and know where you are at any given moment of the day. Would love feedback from anyone who's bounced off traditional trackers for the same reasons.
I ve been building a project called GetVidForge because I kept jumping between different tools for scripts, titles, thumbnails, and keyword research.It s not finished yet, but a few parts already work and I d love feedback from other creators. I m especially interested in what feels useful, what feels unnecessary, and what would make you actually use something like this.If you were testing an early YouTube workflow tool, what would you want it to do first?
We kept seeing support teams answer the same questions over and over again every day.
Traditional chatbots didn t solve the problem well:
too robotic
too limited
too frustrating for customers
So we built RecallAssist. An AI-powered support platform trained on your docs, FAQs, and knowledge base to instantly respond to repetitive customer queries.
For context: I was running cold email campaigns across 4 mailboxes. Every warmup service I tried charged $29/mailbox/month. That's $116/mo before sending a single email.
So I spent 2 weeks building WarmInbox.
It does the same thing the big tools do: Connects to any mailbox via SMTP/IMAP Sends realistic emails to a pool of real accounts daily Opens them, replies 40% of the time Score goes from 0 100 over ~4 weeks
The most interesting part of automation isn't the execution it's the recovery.
I built @SignalForge to be a sovereign execution layer. Instead of a "black box" cloud API, it uses a local observation loop. If you pull the internet plug, it keeps working. If the UI changes, it doesn't crash; it "looks" again, builds a new evidence model, and finds a workaround.
It s like giving your browser a pair of eyes and a local brain that actually learns from its mistakes. No data tax, no cloud leaks. Just pure, deterministic execution on your own hardware. "I m looking for the most hostile, 'un-automatable' web interfaces to test our recovery logic any suggestions?"
Hey everyone, we launched Noa on Product Hunt today.
We built it because keeping life organised still feels way more fragmented than it should be, especially when you re planning with a partner or family.
Been building this for months. Launching on Product Hunt April 29.
Clipo turns one video into clips (ranked by virality), carousels (edit via chat), LinkedIn + Twitter posts (in your voice), subtitles, thumbnails, screenshots and AI images. 12 tools, one upload.
What makes it different: search your video by keyword or describe a visual moment AI finds and clips it. Edit video like a Google Doc. No timeline scrubbing.
