Kefas

There are literally probably 0 actively updated pieces of software you use that aren’t partially made with gen AI for code lol

Vibe Research Fine-tuning a Qwen-8b model to play chess on Replit. Running 3 parallel branches with different experiments and making real progress. It's amazing how far models have come in their ability to do ML (they used to be really bad at it). So now someone with good intuition to guide the process could do interesting ML work, even if they have never done it before.

Codex on iOS pro tip: Dictation mode stays on and continues recording if you background the codex iOS app 1. Open codex app 2. Start dictation 3. Change to my iOS app that I am working on, and play around it for 10 minutes while talking 4. Flip back to codex and send the message My “IDE” doesn’t exist 99% of the time anymore I just play around with my app, and talk, codex listens in the background, i only open codex for 5 seconds to press the send button every few minutes Then I start a new dictation and switch back to my app again And codex makes updates and pushes new builds to my phone as I go, there’s about a 1 second gap as the app closes and reloads each time. So the dev process is 99% me using my app, manifesting ideas, testing how they feel, and watching it change right in front of me. 1% is spent in the IDE. I don’t look at codex, the LLm output or the code. I’m simply experiencing the app and shoving my thoughts and feelings into the machine over and over again

We've reset 5-hour and weekly rate limits for all users.