# Martin Martinez

> Verifying the verifiers: grader & reward-function integrity for RL and evals · formal methods (Lean) · MIT alum

Personal site of Martin Martinez. Each post states its status: verified, exploratory or speculative.
Every post has a raw-markdown copy (links below). Full text of all posts: https://martinmartinez.ai/llms-full.txt

Topics:
- verification: Grader, reward-function and benchmark integrity
- formal: Formal methods: Lean proofs as a reward signal
- harness: Agent harnesses and research tooling
- fabrication: Fabrication: joints, laser cutting, 3D printing

## Posts

- (none published yet)

## Apps

- (none published yet)

## Optional

- [About](https://martinmartinez.ai/about/)
- [RSS](https://martinmartinez.ai/rss.xml)
- [Sitemap](https://martinmartinez.ai/sitemap-index.xml)