Deep Agent

Deploying long horizon agents in production requires purpose-built infrastructure. This guide covers durable execution, memory, HITL, observability, and how deepagents deploy ships it all to production.

An overview of the architectural shift from Shallow Agents (Agent 1.0) to Deep Agents (Agent 2.0) and how to build complex AI agents that can handle multi-step tasks over extended periods.

Deep agents can't be evaluated like simple LLM tasks. After building and testing 4 production agents over the past few months, we learned that evaluating deep agents requires: 1. Bespoke test logic for each datapoint — each test

TLDR: Our coding agent went from Top 30 to Top 5 on Terminal Bench 2.0. We only changed the harness. Here’s our approach to harness engineering (teaser: self-verification & tracing help a lot). The Goal of Harness Engineering The goal of a harness is to mold the inherently spiky intelligence of a model for tasks we care about. Harness Engineering is about systems, you’re building tooling around the model to optimize goals like task performance, token efficiency, latency, etc. Design decisions

tl;dr: Anthropic recently introduced the idea of agent skills. Skills are simply folders containing a SKILL.md file along with any associated files (e.g., documents or scripts) that an agent can discover and load dynamically to perform better at specific tasks. We've added skills support to deepagents-CLI. The Rise of Generalist Agents General purpose agents like Claude Code and Manus have gained widespread adoption. While we might expect generalist agents to use many tools, a surprising tren

Over the past month at LangChain, we shipped four applications on top of the Deep Agents harness: * DeepAgents CLI: a coding agent * LangSmith Assist: an in-app agent to help with various things in LangSmith * Personal Email Assistant: an email assistant that learns from interactions with each user * Agent Builder: a no-code agent building platform powered by meta deep agents Building and shipping these agents meant adding evals for each of them, and we learned a lot along the way! In this