Summary
Noah Provenzano is a graduate student researcher in computer science at Virginia Tech with eight years of hands-on experience building software systems and conducting ML research. He develops and evaluates advanced reasoning and multi-agent inference-time algorithms for LLMs, working with faculty who collaborate with DeepMind, and has prior research on multi-agent reinforcement learning and covert encoding pipelines in PyTorch. Noah has practical production experience as a data and software engineer—running 24/7 dev-ops, building full-stack automation, and delivering monitoring services for clients—plus teaching experience across computer systems and algorithms courses. He combines rigorous academic training in CS and physics with applied engineering skills, and is known informally on GitHub as a "senior scratch engineer," reflecting a playful competence in prototyping and experimentation. Notably, his research has quantified how bits-per-token affect textual similarity while preserving reliable message recovery, demonstrating an attention to both theoretical and empirical evaluation.
8 years of coding experience
2 years of employment as a software developer
Master of Science - MS Computer Science, Master of Science - MS Computer Science at Virginia Tech
Associate's degree Computer Science, Associate's degree Computer Science at New River Community College