Weave Raises $13.5M to Measure Whether AI Coding Works
Weave has raised a $13.5 million Series A for software that attributes engineering work to humans and AI tools and measures the resulting output.
Companies are buying coding agents faster than they can answer a basic question: are these tools actually improving software delivery? Weave wants to provide the measurement layer.
What happened
Weave raised a $13.5 million Series A led by Standard Capital, with Y Combinator, Moonfire, Burst Capital, IrregEx and the Agent Fund participating.
Its engineering-intelligence platform attempts to attribute work to human developers and AI tools, then connect that activity with outputs such as code contributions and delivery progress.
The problem is harder than counting lines of code. An AI agent can generate a large volume of code that later requires review, rewriting or removal. A small human change may solve a much more important problem. Weave therefore needs to interpret contribution patterns without treating raw output as productivity.
Why it matters
Enterprise spending on AI coding tools has increased rapidly, but many technology leaders lack reliable evidence of return on investment. They may know how many licences are active or how many tokens are consumed without knowing whether projects ship faster or software quality improves.
A credible measurement platform could help companies decide where agents are useful, which teams need training and which subscriptions are being wasted. It could also give AI-coding vendors clearer evidence that their products create value.
The bigger picture
Developer measurement carries risks. Metrics can distort behaviour if employees optimise for whatever is counted, and individual rankings may punish people doing essential but less visible work such as architecture, mentoring or incident response.
Weave will need to present its data as context for teams, not a mechanical score of individual worth. Privacy and trust will also matter when the platform analyses detailed engineering activity.
The company says its data covers 20,000 engineers, 500 companies and two million code contributions. Those figures are company-reported and not independently verified enough.
The opportunity is real because AI adoption is creating a new observability problem. The winner will be the platform that measures outcomes without confusing more code with better software.