I had decided to move my tool to a real server. The only problem was I had no idea where to...
#10 The Week I Held My Breath: Moving to a Real Server
Related
I made my agent more capable and it got worse
An ARC Prize 2026 builder's log entry. I taught my agent to solve more kinds of games and the leaderboard score fell from 0.09 to 0.04. The capability was real. The cost of using i...
Why Error Messages Matter More in the Age of AI
Everyone talks about AI writing code. Nobody talks about AI debugging code. Bad error messages are...
How a Transformer Plays Tic-Tac-Toe
An interactive guide to the architecture behind modern language models. Instead of predicting the...