I build software close to where it actually runs, Python for building AI evaluation framework, C/C++/Rust for systems, shell for everything in between, and deep Linux fluency underneath all of it.
Most of my recent work sits at the intersection of evaluation and reliability: how do you actually verify what a language model produces, rather than just trust it. That question runs through my thesis project, my reading list, and the opensource organization I'm building on the side.
Languages
Javascript, Python, C, C++, Rust, Go, Shell
Systems
Linux, CLI tooling, Bash automation
ML / AI
LLM evaluation, Prompting strategy, Execution-based verification, Fine-tuning
Foundations
Data structures and algorithms, Software engineering, Applied AI research
LLM code-generation evaluation framework built from my undergraduate thesis, comparing zero-shot and chain-of-thought prompting with execution-based verification. Working toward an arXiv preprint.
Early-stage technology opensource organization spanning AI systems, firmware, and open-source tooling, designed around safety as a first principle rather than an afterthought.
Chat with models from multiple providers, organize conversations into projects, and bring files and sources into the discussion. NOEHARNESS keeps the next step close to the context that shaped it.