LangChain

Inside Clay's Eval Stack: 300M Agent Runs, One LangSmith Pipeline
How Clay built a 300M-run eval loop.
Decision framework
Before You Build an Internal AI Platform, Decide What the Gateway Should Own
Enterprise AI teams are converging on the same practical question: do you build a custom internal...
Deep dive
Managed Agent Runtimes Save Time. The Overhead Shows Up Later.
Managed agent runtimes are attractive for the same reason most abstractions are: they let teams m...

LangChain Academy Tutors: Learn LangChain with Your Coding Agent
Your IDE is now your teacher.

Inside Toyota's Manufacturing AI: LangChain, LangSmith, and Six Figure ROI
How Toyota uses AI to stop downtime.

Managed Deep Agents - Tools
Unlock AI with custom agent tools

Managed Deep Agents - Intro
Scale agents with managed logic.

Introducing: LangSmith Tuned Evaluators
Automated failure detection is here.
Decision framework
When a Universal AI Agent Stops Working—and Routing Should Take Over
Engineering teams usually start with one universal agent because it is the fastest path to someth...

Building Governed Agents: A Framework for Cost, Control and Compliance
Mastering production-grade agent safety
Decision framework
Autonomous Coding Isn’t the Hard Part. Knowing When to Hand Off Is.
Engineering teams do not need a philosophical debate about whether agents can write code. They ne...
Deep dive
The Loop Is Easy; Verification Is the Hard Part of Code Agents
For founders, eng leads, and investors, the practical question is not whether coding agents can t...

How Madrigal Pharmaceuticals Cut Time to Production From 12 Weeks to 2 with LangChain & LangSmith
Scale AI agents 6x faster.
Deep dive
Your Agent Stack Can Be Observable and Still Unsafe to Ship
The last year of agent tooling has made observability look more mature than it used to. Teams can...
Explainer
Agentic Harnesses Aren’t the Model. They’re the Execution Layer.
If you’ve been following agentic AI, the most useful mental shift is simple: the model is not the...

Building Deep Agents and Deploying in Production
Stop building agents, start building harnesses.

Autonomous Agent Improvement with LangSmith Engine | New LangChain Academy Course
Scale agents using automated cycles.