Velocity Lab
AIPrototype

Velocity Lab

Team Guntinun Sawatvong

The big news

Developers get performance gains they can trust. Velocity is an evidence-backed AI agent that analyses a public Git repository, finds a measurable bottleneck and proposes a focused patch. It then verifies the change with tests and repeated benchmarks against the original baseline, filtering out improvements that are just noise, extra memory use or new bugs. Only then does it ask a human to accept or reject the result. It's AI optimisation that shows its receipts.

Watch the demo

What it does

An AI performance-engineering agent that builds, tests and benchmarks proposed optimisations against a baseline before presenting them for human review. Built with Python, TypeScript and C++, packaged with Docker.

Try it yourself

U-tags

Tap a tag to find similar projects

Meet the team

GSGuntinun Sawatvong

This is an independent hackathon project. Some features may be prototypes. Please do your own research before relying on it.

Projects are independent hackathon entries by their builders. They aren't official U-Topia products, and some features may be prototypes or simulations.