Weave

Weave is an engineering analytics platform that uses LLMs and machine learning models to measure how much work engineers and AI agents produce, how good it is, and how AI contributed. It is sold to engineering organizations that want to connect AI adoption to output, quality and cost. The platform combines AI usage metrics, DORA, SPACE and developer feedback to show where work gets stuck. Dashboards compare AI assisted merges, team output against a baseline and code quality by engineer. A token intelligence view links AI spend to output and benchmarks it against other organizations. Wooly is an AI agent that analyzes the data and suggests improvements, available in the app or through an MCP server. Weave also offers an open source Router for coding agents, which classifies each prompt and sends it to a model chosen by cost, speed and quality. It installs with a single npx command and supports Anthropic, OpenAI and Google clients. The vendor lists SOC2 Type II compliance, and Enterprise options include GitHub Enterprise support and custom terms. Published platform plans are Starter at $0, for basic usage or teams under five, Pro at $50 per engineer per month billed monthly, and a custom priced Enterprise plan. The pricing page also shows a yearly billing option with a discount of 16.67 percent.