Case Study: Pydantic auto-approves 91% of PRs with Macroscope

Pydantic auto-approves 91% of PRs with Macroscope

Pydantic, the company behind the most depended-on library in Python, faced a challenge where its coding agents were generating pull requests faster than its human engineers could review them. This created a significant bottleneck in their development process. To address this, Pydantic chose Macroscope and its Approvability product to automate the review and merging of safe changes.

By implementing Macroscope, Pydantic was able to auto-approve roughly 91% of its pull requests without human intervention, escalating only the changes that required human judgment. This solution, combined with Macroscope's Murmur agent orchestration platform, increased their weekly PR output by 4x, from 100-150 to approximately 500. The efficiency gains were substantial, allowing the team to ship a full epic of work in an afternoon instead of 15 hours.

Pydantic