LANGCHAIN

Harrison Chase was an engineer at Robust Intelligence, a machine learning startup. In October 2022 he open-sourced a small side project called LangChain.

It was a toolkit for chaining language model calls into real applications. A month later ChatGPT launched and the whole world wanted to build with AI.

LangChain was already sitting right there. The repo exploded.

Chase co-founded the company with Ankush Gola in early 2023 and raised money almost immediately.

The LangChain framework is free and open-source. Developers use it to connect language models to data, tools, and each other.

The money comes from LangSmith. That is a paid platform for testing, watching, and debugging AI apps once they go live.

The free tool pulls developers in. The paid tool keeps the companies that got serious.

Classic open-source-to-enterprise playbook.

LangChain is the core open-source framework for building apps on top of language models. LangGraph handles AI agents that need to loop, branch and make decisions.

Both hit version 1.0 in October 2025, and together they pull about 90 million downloads a month. LangSmith is the paid platform for testing, monitoring and now deploying those agents in production (the deployment piece used to be called LangGraph Platform).

Agent Builder, a no-code way to make agents, launched in preview in late 2025. The open-source pieces get the attention.

LangSmith pays the bills.

The growth hack was timing plus generosity. LangChain gave away the framework right as millions of developers needed one.

Every AI tutorial, hackathon and demo in 2023 seemed to run on it. That built a giant community fast.

Then LangGraph arrived for building AI agents, and LangSmith arrived to charge the serious users. By October 2025 the company said 35% of the Fortune 500 used its services.

Ride the wave, then sell the surfboard.

Being early is not the same as being safe. LangChain got a reputation for messy code and too many layers of abstraction.

Plenty of developers ripped it out and built their own tools instead. Model makers like OpenAI also keep adding features that overlap with what LangChain does.

The hard part is staying essential when the models themselves get smarter every few months.

Benchmark led a $10 million seed in April 2023. Sequoia Capital led a $25 million Series A at a $200 million valuation, announced in February 2024, with Lux Capital and Conviction joining.

In October 2025 IVP led a $125 million Series B at a $1.25 billion valuation, with Sequoia, Benchmark, CapitalG, Sapphire Ventures and Amplify back in, plus strategic checks from companies like ServiceNow, Workday, Cisco, Datadog and Databricks. Documented rounds add up to about $160 million.