antirez: Good devs struggling with GPT 6 Astra need a different skill set, not a better model
mitsuhiko · x · 2026-09-27
antirez responds to strong programmers reporting poor results with GPT 6 Astra: the explanation isn't model quality but that good programming now requires a different skill set, overlapping only partly with the old one. mitsuhiko says he's in that adjustment phase, currently using Astra to discuss problems and Opus 5.5 for implementation, and finds the latter yields fewer weird results.
More from coding & agent
- TypeSafe ships official skill teaching Claude Code to batch Jev calls, cutting bills 12.2x — anselm · 2026-09-27
- World's fastest panel QR factorization on B200: how a GPU MODE contestant cracked chained dependencies — A_K_Nain · 2026-09-27
- Graph Engineering: Author Finds Agent's Self-Checking Was Lying to Him — blaizedsouza · 2026-09-27
- One prompt in OpenCode builds a full 3D solar system observatory with themes — JaynitMakwana · 2026-09-27
- Stop building agents as a straight line: the 90-second test to cut fake dependencies — blaizedsouza · 2026-09-27
- Digital Apprentice framework: agents earn autonomy through per-skill authorization gates — blaizedsouza · 2026-09-27