ML/AI Research Engineer · AI Systems Researcher · Open-Source Builder

I research and build agentic systems, programming languages, compilers,
and AI-native developer infrastructure.

A verified recursive agent runtime for software engineering. It explores competing implementations, verifies them against explicit acceptance contracts, promotes proven changes, and learns from evidence.

An experimental statically typed native programming language designed for humans and coding agents. The research explores concise syntax, native performance, explicit effects and capabilities, semantic representations, and AI-native tooling.

Agentic Systems · ML Systems · LLM Infrastructure · Programming Languages · Compilers · Semantic Code Intelligence · AI for Software Engineering

• Verified multi-agent software development
• Semantic representations for AI coding agents
• Native performance without explicit lifetime syntax
• AI-native programming language design
• Agent context compression and semantic codebases

These are technical working notes: concrete research questions, system designs, evaluation criteria, and open problems—not claims of peer review.

I am interested in systems where AI does more than generate text: agents that reason over codebases, verify their own work, operate under explicit constraints, and improve through measurable evidence.

My current work also explores programming models designed for both human clarity and machine understanding.

Building systems for the next generation of software development.