AI infrastructure for a 600+ engineer organization.

• Delivered an org-wide AI skills marketplace for Claude Code, Cursor, and Copilot, adopted across all engineering stacks.

• Built a LangGraph-based agentic QA platform with human-in-the-loop approval and checkpointed resumability.

• Drove an AI-led E2E test migration (≈800 tests), improving developer productivity ≈2.5x.

• Founded and lead a cross-functional AI SDLC guild that partners with teams to identify workflow friction and convert repeated patterns into reusable skills and hooks.

Open-source platform for deterministic AI coding workflows.

• Built Agent Validator, a configurable "feedback loop" runner pairing static checks with cross-agent AI reviews for autonomous coding workflows.

• Designed Agent Skills, a multi-agent plugin for spec- and test-driven development - proposal, design, TDD implementation, and PR shepherding.

• Building Agent Runner, a declarative orchestrator for multi-agent coding workflows.

• Author of the Codagent newsletter on agent harness engineering.

Core contributor in the development of Knewton’s consumer-facing adaptive learning product. Led a team of five.

Helped to create Mindspark’s suite of social and entertainment websites as well as various internal platforms and solutions. Led a small team.

Delivered custom software development solutions for clients in the capital markets industry