Harrison Chase's North Star: agents that improve via model, harness, and context updates
hwchase17 · x · 2026-11-09
LangChain founder Harrison Chase's Interrupt London talk lays out a "North Star" vision of AI agents that improve over time, decomposing an agent into three core parts: the Model (reasoning engine), the Harness (surrounding control system), and Context (memory and knowledge).
Three dimensions of improvement:
- Model: update weights via Supervised Fine-Tuning (SFT) for simpler tasks or Reinforcement Learning (RL) for complex ones.
- Harness: refine code, prompts, and tools; LangSmith Engine automates this by monitoring traces to spot issues and propose fixes.
- Context: the easiest dimension to update — evolving memory at the user or organizational level.
He argues traces and evals are the critical foundation underlying all improvement.
More from coding & agent
- Hark Labs' model + agent + VM experience beats Codex, says early reviewer — Scobleizer · 2026-11-10
- Dev rebuilds Pokémon Red in 2.5D from his own ROM cartridge using an agentic workflow — NathanWilbanks_ · 2026-11-10
- Same prompt shootout: Claude vs ChatGPT vs Figma Make on an e-commerce landing page — Tegadesigns · 2026-11-10
- Publishing an MCP server to the official registry: the 403 gotcha and immutable versions — srikanth_builds · 2026-11-10
- Sonner creator ships /gestures skill packing rubber-banding and momentum drags for apps — floguo · 2026-11-10
- NotiBuddy: Open-Source MCP Tool Lets Coding Agents Ping Your iPhone via Lockscreen — dbenc · 2026-11-10