I build AI systems end to end—computer vision, generative AI, agent tooling, and the full-stack systems around them—with reliable workflows and clear failure handling in production.

• At DaoAI Robotics, I work on production video analytics and generative AI platforms, spanning workflow validation, multi-object tracking, multimodal product workflows, and persistent asynchronous job orchestration.
• In open source, I build AI agents and developer tools whose work can be inspected, reproduced, and verified.

• Loop Agent — Autonomous software delivery from a product brief to a verified implementation, with reviewable artifacts, Git changes, and a replayable receipt.
• Argus — Evidence-first MCP QA for web and native macOS apps, with clean-state reproduction and verified bug replay in CI.
• Sibyl — Web research for AI agents with keyless multi-source retrieval, passage-level evidence ranking, source clustering, and citation-backed reports.
• Scalable Starter — A production-shaped Next.js and FastAPI foundation with Postgres, Redis-backed workers, observability, and Kubernetes deployment.
• Agent Duet — Local Codex and Claude Code collaboration with independent review, finite rounds, machine checks, and guarded Git application.

• Languages: Python, TypeScript, JavaScript, SQL
• AI / ML / CV: LLM and VLM APIs, agent and tool orchestration, PyTorch, NumPy, SciPy, OpenCV, multi-object tracking, OCR
• Backend & Data: FastAPI, Celery, asyncio, SQLAlchemy, Pydantic, Redis, PostgreSQL, WebSockets
• Frontend & Media: React, Next.js, Tailwind CSS, FFmpeg
• Cloud & Delivery: AWS, Docker, Kubernetes, Helm, GitHub Actions
• Testing: pytest, Playwright, Vitest

• MS in Computer Engineering — Northwestern University
• BS in Computer Engineering — University of Illinois Urbana-Champaign, Dean's List
• AWS Certified Solutions Architect – Associate and AWS Certified Cloud Practitioner