Skip to content
View stanleyngugi's full-sized avatar

Block or report stanleyngugi

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
stanleyngugi/README.md

Stanley Ngugi

Independent AI researcher working on post-training with reinforcement learning, RL environments, and formal verification.

Current work

Instead of rewarding C code for passing a test suite, Formally Verified C rewards a model only when Frama-C proves that its implementation satisfies a fixed ACSL specification, including runtime-safety obligations. The public alpha includes 64 tasks, an isolated judge, adversarial negative controls, and reproducible proof evidence. Technical article.

An RL environment for mathematical reasoning that does not rely on hidden answer keys. Models return an answer or a complete finite certificate, and a frozen problem specification with a Lean-backed checker determines the reward. Technical article.

The verification engine behind MathCheck RL. It turns bounded mathematical specifications into Lean checks for exact answers and complete finite relations, returns explicit failure reasons, and can evaluate untrusted model output in isolation. Technical article.

Selected writing

Earlier research

These preprints are part of my earlier research; my present work centers on RL environments and formal verification.

Writing

I publish research notes and technical essays at stanleyngugi.netlify.app.

Research questions, criticism, and collaboration proposals are welcome through the contact links on my website.

Pinned Loading

  1. formally-verified-code-rl formally-verified-code-rl Public

    RL environments for generating code whose correctness is checked by formal verification; first release: C + ACSL + Frama-C.

    Python

  2. mathcheck-rl mathcheck-rl Public

    Answer-key-free bounded-math tasks and Lean-checked rewards for reinforcement learning.

    Python

  3. ai-proof-grammars ai-proof-grammars Public

    Grammars for AI-generated mathematical proof steps. Experiments with Lean, saved model outputs, and reproducible analyses.

    Python

  4. mathcheck-engine mathcheck-engine Public

    Bounded mathematical answers checked by generated Lean 4 programs.

    Python