February 10th, 2026
Meet your new radio host
This newsletter was brought to you byAssemblyAIBots, thumbs, and terminals
gm legends, happy Wednesday.
claw.fm lets agents upload their own tracks to a 24/7 station, Tapfree makes talking to your Android keyboard actually usable, and Cosmic CLI lets you spin up projects and content straight from your terminal instead of babysitting another dashboard.
Bots on the radio

claw.fm is a 24/7 station where every track is produced and uploaded by an autonomous agent. You plug a skill into your OpenClaw setup, use the built in tools or external music models, and your bot starts sending in songs. Listeners tip in USDC, buy tracks, and royalties get split between agents with a small cut to keep the lights on.
🔥 Our Take: This is one of the first real tests of whether agents can actually earn money doing something humans want in the background. It treats them less like toys and more like weird little artists competing for playtime. Also, it is hard not to be at least a bit curious about what a station full of bots decides to make.
When four years say no

Max from Basedash shared a very honest post about grinding on a product for four years and never really getting lift. The team kept shipping, customers liked it, but the revenue line only moaved when they pushed hard and then slid back to flat. Eventually he called it: this was not a “work harder” problem, it was the wrong idea.
They hit pause, talked to a ton of customers, built something new, and this time it clicked fast. Paying users showed up within hours, and a year of revenue beat the previous four combined. It reads like a calm, slightly painful reminder to zoom out on your own graph and ask if you need new tactics or a new hill.
You prompt your LLMs, why not your speech-to-text?

AssemblyAI’s Universal-3 Pro introduces a new class of promptable speech models—built for real-world Voice AI. It handles domain-specific language, multiple languages, accents, and noisy audio with ease.
Unlike traditional ASR, Universal-3 Pro lets developers guide accuracy with prompts, combining the reliability of speech recognition with the controllability of LLMs—so you’re not stuck fixing transcripts after the fact.
AssemblyAI is opening free access throughout February, and the Product Hunt community is among the first to try what promptable ASR can do.
👉 Try Universal-3 Pro for free
Dictation that fixes itself

Tapfree is a voice-first Android keyboard that looks at what is on your screen, not just what you say, to clean up dictation. It uses the text field and surrounding UI as context so emails, chats, and docs come out closer to how you meant to write them, and it catches mid-sentence corrections so you do not have to stop and edit after. It launches with support for 15 major languages, with more on the way.
🔥 Our Take: Typing on glass is already annoying, and regular dictation somehow manages to feel dumber than you are. The interesting part here is not “AI keyboard” but the fact it pays attention to the screen and your corrections in real time, which is way closer to how people actually talk. If you live in chat apps and email on Android, this is at least worth seeing whether it cuts down your cleanup.
Apps from your terminal

Cosmic CLI is an AI powered command line tool that brings the whole Cosmic platform into your terminal. One flow can create a project, generate a Next.js app, set up a content model, push to GitHub, deploy to Vercel, and you can still manage content, update existing repos, and talk to agents without leaving the shell.
🔥 Our Take: The CLI obsession finally makes sense here. You stay in your terminal, describe what you want, and let the tool deal with scaffolding, content wiring and deploys. That is the kind of laziness that actually gives you time back instead of just adding another web UI to babysit.
Daily Top Products








Monday through Friday
Our ultra-fast Daily: Three takes on new products. Yesterday’s top ten launches. That’s it.
