How to apply context engineering

Agents need context (e.g., instructions, external knowledge, tool feedback) to perform tasks. Context engineering is the art and science of filling the context window with just the right information at each step of an agent’s trajectory. There are many context engineering approaches, and Drew Breunig recently summarized six of the most common in his recent blog post titled "How To Fix Your Context." Here, we show how to apply each of these methods from scratch using LangGraph.

Learn to build with LangGraph on LangChain Academy:
https://academy.langchain.com/collections/quickstart/?utm_medium=social&utm_source=youtube&utm_campaign=q4-2025_youtube-academy-links_aw

Repo:
https://github.com/langchain-ai/how_to_fix_your_context

Blog:
https://www.dbreunig.com/2025/06/26/how-to-fix-your-context.html

Slides:
https://docs.google.com/presentation/d/1UqJ9xB0QQuyQbILRxgsSyIgobtu3GpzoRrHrUwj7cCI/edit?usp=sharing