Blacksmith, a startup that runs software testing infrastructure for engineering teams, has raised $66 million in a Series C round led by Sequoia Capital. The round values the company at $1 billion. Premji Invest came in as a new investor, and existing backers Peak XV Partners, GV and Y Combinator also participated.
The funding takes Blacksmith’s total raise to about $125 million. It closed a $45 million Series B led by Peak XV in March 2026 at a $550 million valuation. That round was announced only in August. The company has doubled its valuation in roughly seven months.
Read More: AI Hiring Startup Firstwork Raises $15 Mn Series A Led By Earlybird
Read More: Startup Claims It Can Make Gold From Mercury, Physicists Want Proof
What Blacksmith actually sells
Blacksmith provides infrastructure for continuous integration. That is the step where automated checks run against every code change before it is merged. Most teams rent cloud machines for this work. They configure those machines, scale them and pay for them by the minute.
Blacksmith replaces that with a managed service. The company says it completes CI jobs twice as fast and at about 60% lower cost. More than 10,000 companies use the platform, according to Blacksmith, including Ashby, Granola, Handshake and LangChain. It runs close to 10 megawatts of computing capacity across data centres in the United States and Europe.
Why AI coding agents created the demand
The pitch follows directly from how software is now written. A developer who once opened three or four pull requests a day can now generate dozens with an AI coding agent. Every one of those changes still has to be tested before it reaches production.
“Their CI infrastructure can’t keep up,” cofounder and chief executive Aditya Jayaprakash said of the teams Blacksmith sells to. The usage numbers explain the urgency. The company processed five million CI jobs a week in January 2026. It now handles more than 90 million a week, it said. Annual recurring revenue grew 17 times over the same period.
The picks-and-shovels trade in AI coding
This is one of the cleaner ways to profit from AI coding tools without competing with them. Cursor, Claude Code and GitHub Copilot are fighting for the developer’s attention. Blacksmith sells the capacity that catches what those tools produce.
Sequoia partner Isaiah Boone said the company recognised early that AI-generated code would need different testing infrastructure. He cited the founding team’s systems expertise. The founders met at the University of Waterloo. Aayush Shah and Aditya Maru were engineers at Cockroach Labs, and Jayaprakash worked at Faire.
The public proof points matter more than the logos. Joaquin Borggio, an infrastructure engineer at LangChain, said Blacksmith cut the company’s runner costs by roughly 40%. That figure excludes the engineering time saved. A customer who quantifies savings in public is a stronger signal than a name on a landing page.
The capital intensity is the risk
Testing infrastructure is not an asset-light software business. Blacksmith manages nearly 10 megawatts of compute today. It plans to expand that at least fivefold in 2027. Compute is bought up front, so margins depend on utilisation staying high as capacity grows.
That is where the comparison with pure software companies breaks down. A 17-fold rise in annual recurring revenue sounds extraordinary. It is also measured against a small base, and the company has not disclosed the absolute figure. At a $1 billion valuation, investors are pricing years of growth that has not been booked yet.
Competition is the second risk. GitHub Actions, Buildkite and a cluster of newer CI vendors chase the same workloads. The large cloud providers can bundle testing capacity into deals Blacksmith cannot match on price alone. There is a structural question too. If AI coding agents keep improving, some of the testing load they create may be filtered before it ever reaches a pipeline.
What to watch
Three numbers will decide whether the $1 billion holds. The first is absolute annual recurring revenue. Disclosing it would turn a 17-fold multiple into a growth rate that can be judged.
The second is gross margin as compute capacity goes up fivefold. A managed infrastructure business that cannot hold margin at scale is a services company with software branding. The third is customer mix. A platform that grew on AI-native startups has to prove it can win enterprise engineering teams on long contracts.
What Blacksmith has shown is that the boring layer of software delivery became a growth market the moment AI made code cheap to write. The testing bottleneck is real, and the company is one of the few selling directly into it.