Note This is a shortened version of the original writing on kig.re — my personal blog. Here is...
Home-made Evals Just Taste Better
Related
Automating AI Skill Updates with Dependabot and APM
Use Dependabot and Agent Package Manager to receive shared AI skill updates as PRs and keep installed skill files in sync.
Economía de contexto — el 99.2 %
El 99.2 % del coste de un agente es input, no output. Medido sobre dos generaciones del mismo panel: el total se mueve 28.19 % entre corridas, la proporción se mueve 0.06 puntos. P...
I tried to break my AI song detector with covers, live recordings, lo-fi and Suno v6
In October I tested 8 AI song detectors on 65 songs. The winner caught 23 of 25 AI songs and flagged...