From Traces to Insights: Understanding Agent Behavior at Scale Visibility is the easiest piece. The hard part is analyzing and understanding what you’re observing. I’ve spoken to teams recording 100k+ traces every
In software, the code documents the app. In AI, the traces do. TL;DR * In traditional software, you read the code to understand what the app does - the decision logic lives in your codebase * In AI
Agent Frameworks, Runtimes, and Harnesses- oh my! There are few different open source packages we maintain: LangChain and LangGraph being the biggest ones, but DeepAgents being an increasingly popular one. I’ve
Reflections on Three Years of Building LangChain by Harrison Chase Almost exactly 3 years ago, I pushed the first lines of code to langchain as an open source package. There was no
Not Another Workflow Builder By Harrison Chase One of the most common requests we’ve gotten from day zero of LangChain has been a visual workflow builder. We never
Deep Agents Using an LLM to call tools in a loop is the simplest form of an agent. This architecture, however, can yield agents that are “shallow”
The rise of "context engineering" Header image from Dex Horthy on Twitter. Context engineering is building dynamic systems to provide the right information and tools in the right format such
How and when to build multi-agent systems Late last week two great blog posts were released with seemingly opposite titles. “Don’t Build Multi-Agents” by the Cognition team, and “How we built
The Hidden Metric That Determines AI Product Success Co-authored by Assaf Elovic and Harrison Chase. You can also find a version of this post published on Assaf's Medium. Why do some
How to think about agent frameworks TL;DR: * The hard part of building reliable agentic systems is making sure the LLM has the appropriate context at each step. This includes both
How do I speed up my AI agent? I get this question a bunch. Developers generally first spend time getting the agent to work, but then they turn their attention to speed and
MCP: Flash in the Pan or Future Standard? Model Context Protocol (MCP) is creating quite the stir on Twitter – but is it actually useful, or just noise? In this back and forth, Harrison
