Selected work

A few projects that show how I think — problem, constraints, architecture, tradeoffs, and what I’d change next. Each links to code, writing, or a demo.

AI Workflow Kit

Creator & sole maintainer

A portable skill format and integration-abstraction layer — one markdown skill runs identically across Claude Code, Cursor, and Codex and binds to any calendar/notes/chat vendor at setup, not in code. The shipped library automates the recurring non-coding parts of an engineer’s day.

Agentic workflowsDeveloper toolingTool use & orchestrationIntegration design

Long-Running Coding Agents: A 6,000-Line Migration

Planner & architect (human-in-the-loop); AI agent as executor

Completed a 6,000-line Python-to-TypeScript migration with an AI coding agent by treating the model as a stateless executor and myself as the planner — using an external progress file to hold state across sessions.

Agentic workflowsLong-running executionHuman-in-the-loop systemsEvaluation & feedback loops

Adaptive Prompt Selection & Evaluation

Designer & author of the approach

A design for moving beyond static prompts: retrieve candidate prompts by context, select with a multi-armed bandit, handle cold start, and close the loop with multidimensional evaluation and reward design.

Retrieval & RAGEvaluation & feedback loopsModel integration & routingApplied AI

MCP Tooling & Tool-Boundary Design

Author of original servers; adapter/maintainer of forks

Model Context Protocol servers built on one principle — narrow, scoped tool surfaces over generic passthroughs — so AI assistants can safely act on real systems. The harder work is the tool-boundary design: scope, error handling, and where a human stays in the loop.

Tool use & orchestrationIntegration designDeveloper toolingHuman-in-the-loop systems