tl;dr: Weave uses AI to measure software engineering work in the AI era.
It shows how much work is being done, how much is done by AI and where you can utilize AI tools to improve your engineering process.
The Problem
Historically, it’s been impossible to measure engineering so leaders are left flying blind. Engineering teams rely on gut feel or shoddy metrics to try to get a handle on what’s going on and where to improve.
That's why we built Weave.
The Solution
Weave uses AI to measure engineering work. We run LLMs and our own models on every PR and review, analyzing both output and quality. We integrate with all AI coding tools (Claude, Cursor, etc.) and have PR level code attribution to determine what was written by AI, what wasn’t, and what should’ve been.
We summarize this data and insights in dashboards. Teams like Reducto, Superpower and Laurel use Weave to ship 16% more, just 2 months after using Weave.
AI Usage
We show you who your top AI performers are (so they can share best practices), how the team stacks up against its competitors and the real financial return of your AI investments.
AI Insights
We keep you up to date with the best practices and tools used by the most forward-thinking AI teams.
Measuring Code Output
This isn't a line of code calculator, this is an actual estimate of our key metric, a Weave Hour: "How long would it take an experienced engineer to make this change?"
We can also tell you how much time each engineer is spending on code review and how helpful their reviews are:
Individual
See output across individuals to identify top performers, see how new hires are ramping and if anyone is blocked.
Our Story
Adam’s background is in operations and sales. He led organizations of 100+ people and created an internal tool to measure performance and help individuals identify their weak spots. This is common practice in revenue teams and he wants to bring it to engineering.
Andrew was employee #1 at Causal. He saw firsthand how subjective engineering management is and how hard it is to scale a high-performing engineering team.
We met at Causal, where Adam was hired to run the sales and customer success function. We got to talking about the big difference between how our two departments worked, and the rest is history.
Early Use Cases
- Reducto, Superpower, and Posthog have used Weave to see increases in output correlated with AI usage (16% on average)
- A manager noticed that code review quality has the highest correlation to output. He reset code review standards and team output went up by 15%
- Smaller engineering teams (~5 engineers) are using the platform daily in stand-ups
Our Ask