We built a structured entity graph for an AI agent. Then we removed most of itblog.getcassis.com·2 pts·matthieu_bl·2
Anthropic appears to be A/B testing reduced effort levels in Claude Codetwitter.com·216 pts·matthieu_bl·191
Context engineering for analytics agents: six months of building and rebuildingblog.getcassis.com·1 pts·matthieu_bl·0
Show HN: A skill to audit your dbt project for what an AI agent will get wronggithub.com·3 pts·matthieu_bl·0
Snap lays off 16% of workers, citing 'rapid advancements' in AIbusinessinsider.com·12 pts·matthieu_bl·7
OpenAI Codex Telepathy feature flag: sidecar for passive screen-context memoriesgithub.com·1 pts·matthieu_bl·0
Claude Managed Agents: everything you need to build and deploy agents at scaletwitter.com·1 pts·matthieu_bl·0