Every talk: turn "this answer feels better" into tests for models and agents
every · x · 2026-11-05
Every's free Thesis: 2027 talk by @hammermt shows how to convert the instinct "this answer feels better" into reusable tests by combining human judgment, model-based checks, and code to compare models, improve agents, and catch regressions. The full-day event also features speakers from OpenAI, Sequoia, Notion, The Atlantic, The Browser Company, and more, with $20,000 in partner credits.
More from coding & agent
- Warp terminal praised as a fast, memory-efficient open-source ADE with agent orchestration — vikvang1 · 2026-11-06
- One prompt turns Opus 5.5 into a task graph so agents only need approval on the last step — CurieuxExplorer · 2026-11-06
- Pinokio 8.3.7 ships security-first architecture after eMudhra certificate breach — cocktailpeanut · 2026-11-06
- Cryptographer Matthew Green: Attestable LLMs could let agents verify shared goals — matthew_d_green · 2026-11-06
- Dev complains Claude Code quality has tanked: same models, shockingly worse results — IanArawjo · 2026-11-06
- Braintrust exposes its MCP server in the bt CLI: why not both MCP and CLI — indragie · 2026-11-06