Weave
VisitEngineering and token intelligence to optimize the SDLC. Weave shows you the ROI of every AI dollar, benchmarked against thousands of orgs.
Weave connects to your GitHub and your AI coding tools and turns both into one picture of how your engineering org actually works.
On the output side, Weave normalizes work into comparable units so you can see output per engineer, per team, and per component, then benchmark it against thousands of comparable orgs instead of against your own gut feel.
On the cost side, Weave captures token spend from Claude Code, Codex, Cursor, and any AI gateway, then breaks it down by engineer and by model. Line by line AI attribution shows how much code came from a person and how much came from a tool, so cost per unit of output becomes a number you track rather than a number you guess at.
Weave also includes Wooly, a research agent with access to your data and our docs, an MCP server and API for pulling metrics into your own workflows, and a router that sends simpler prompts to cheaper models and reports exactly what you saved.
