Current focus
Production agent systems.
I work on AI agents as shipped software: product surfaces, tool use, runtime behavior, evaluation, and debugging loops.
- Systems
- Production agents
Build agent systems around tool use, memory, evaluation, observability, and recovery paths.
- Runtime
- Product + infrastructure
Connect model behavior to product UX, orchestration, debugging surfaces, and release workflows.
- Measure
- Research to product
Use evaluation and production feedback to move promising behavior into shipped systems.
Current work
Agent platforms, from interface to runtime.
User-facing agents with the runtime behind them.
Tool routing, memory, orchestration, evaluation harnesses, and debugging views for behavior that has to survive real use.
Evaluation as part of the product loop.
Instrument failure modes, compare behavior, and route high-signal cases back into model, tooling, and UX decisions.
ML systems and infrastructure experience.
Distributed systems, traffic engineering, applied ML, and product engineering, with a bias toward reliability and legibility.
Selected systems
Recent work in agents, ML systems, and infrastructure.
Production agent platforms
Product and systems work on production AI agents, spanning user surfaces, tool use, runtime behavior, and evaluation.
Agent harnesses and eval loops
Harnesses that let agents use tools, retain context, recover from failures, and improve through measured feedback.
ML systems and network intelligence
Built prediction, optimization, automation, and applied ML systems for infrastructure at production scale.
Research
Selected public research and engineering work.
Background
Recent roles and training.
Meta TBD Lab
AI Research Scientist working on AI agent products and systems.
Netflix
Senior Research Engineer on ML systems and production infrastructure.
UIUC + Tsinghua
Ph.D. Transportation, M.S. Applied Mathematics, and B.E. Civil Engineering and Economics.
Contact