AI codes too fast, testing can't keep up: Blacksmith raises $66 million, valued at $1 billion

Beating AI News Flash: AI code testing company Blacksmith has announced the completion of a $66 million Series C funding round, led by Sequoia Capital, at a valuation of $1 billion. The company was founded in 2024, and its valuation was only $550 million when it completed its Series B funding round in March this year.

Blacksmith handles the testing work after AI writes code. Tools such as Claude Code and Codex can now quickly generate large amounts of code, but this code cannot go directly into production and must still go through compilation and testing to check for errors. The faster code is generated, the more testing tasks are required, and existing systems are beginning to fall behind. Blacksmith specifically provides servers to run these tests, so developers do not have to wait a long time for results.

The code compilation, testing, and other tasks Blacksmith processes each week have increased from 5 million in January this year to more than 90 million, a roughly 18-fold increase. During the same period, the company's annual recurring revenue grew 17-fold, with more than 10,000 customers, including LangChain and Granola. Blacksmith says its service can on average double testing speed and reduce costs by 60%.

Unlike traditional solutions that mainly rely on renting cloud servers, Blacksmith purchases its own servers and deploys them in data centers. Most of the funding from this round will also continue to be used to purchase computing power. The company plans to expand its computing power scale by at least 5-fold by 2027.