LangChain founder asks: is agent-level memory just 'skills' with a trace-updating process?
hwchase17 · x · 2026-11-03
Harrison Chase, creator of LangChain, poses a conceptual question: is org/agent-level memory essentially just "skills," assuming there's a process for updating skills from agent traces and history? He invites arguments for and against, touching on the boundary between memory and skill consolidation in current agent architecture debates.
More from coding & agent
- AI coding is 10–20x faster for the first 1–2 months, then complexity drags it back down — LadyAshBorg · 2026-11-03
- KAIST's ReTrace Demystifies Computer-Use Agent Recovery, Finds Reflection and Rollback Fragile — kaist-ai · 2026-11-03
- TypeGPU + three.js piano visualizer remake: dev says the slick API is agent-friendly — nptacek · 2026-11-03
- AI swarm of "neural bees" communicates in neuralese, showcasing agent ecologies — Sauers_ · 2026-11-03
- Human mops while their paid AI agent renders an After Effects project solo — ZeroStateReflex · 2026-11-03
- 9 GitHub skills repos that turn Claude into a real work assistant — nikola_mr64990 · 2026-11-03