Harrison Chase, founder of LangChain, reflects on the company's three-year journey from a side-project open source Python package to a $1.25 billion company. The post recounts the launches of LangSmith (observability/evals) and LangGraph (a controllable agent runtime), and announces LangChain 1.0 - a ground-up rewrite focused on a core tool-calling loop, a new middleware concept for context engineering, and built on LangGraph's runtime. It also announces a $125 million Series B funding round and the evolution of LangSmith into a broader 'agent engineering platform' that now includes deployments.
Table of contents
Starting as a side projectForming a companyLaunching LangSmithLaunching LangGraphRevisiting LangChainEvolving LangSmith into the Agent Engineering PlatformAgents of the futureQuestions this post answers
What is new in LangChain 1.0 compared to earlier versions?
LangChain 1.0 is a ground-up rewrite focused on the core tool-calling loop that defines agents, adding a new middleware concept for controlling the context engineering lifecycle, and building on the LangGraph runtime for streaming, durable execution, and human-in-the-loop support. Older 0.x versions remain available as langchain-classic for continued support. Track major framework rewrites like this one as you plan an agent stack migration on daily.dev.
What is the difference between LangChain and LangGraph?
LangChain is the higher-level framework designed to make it easy to get started building agents, while LangGraph is the underlying production runtime providing streaming, statefulness, human-in-the-loop, and durable execution. LangChain 1.0 is now built directly on top of LangGraph rather than existing as a separate, less controllable abstraction. Compare agent framework tradeoffs like this one before picking a stack on daily.dev.
How much funding did LangChain raise and at what valuation?
LangChain raised a $125 million funding round at a $1.25 billion valuation, announced alongside the LangChain 1.0 release. The funding is earmarked for expanding LangSmith into a broader agent engineering platform and growing open source contributions. Follow funding and product shifts like this to gauge which agent tools are gaining momentum, on daily.dev.