Vibeleaderboard
← All Intel
Intel / video

Context Engineering Our Way to Long-Horizon Agents: LangChain’s Harrison Chase

Source
youtube.com
Author
Sequoia Capital
Date

Context

Harrison Chase of LangChain says in a Sequoia Capital interview that agents, which run a language model in a loop that picks its own next step, are finally starting to work for long stretches, mostly in coding. Chase credits better models together with better harnesses, the code wrapped around a model.

Chase separates a framework, which offers swappable parts such as tools and memory without dictating choices, from a , which is opinionated. In this description, LangChain's Deep Agents harness includes a planning tool by default, compacts history when the fills, and gives agents file system tools. Models are also being trained on file-system-style , Chase says, so harness and model have evolved together. Because an agent's context at step 14 depends on 13 earlier steps, traces are needed to see what the model was given.

Terms in this piece · Glossary
  • agent harness — The scaffolding around a model that turns it into a working agent — the loop, the tools it can call, and the rules for when to stop.
  • context window — The maximum amount of text a model can consider at once — its working memory for the current conversation or task.
  • tool use — A model's ability to call external functions — run code, search the web, edit files — instead of only generating text.
  • context engineering — Deciding what an agent sees at each step — the discipline that replaced prompt engineering once context windows got large enough to fill badly.
Read the source www.youtube.com
More from Sequoia Capital
Recommended reads
Comments

Checking sign-in…

Loading comments…