evaluated GPT-6 Astra and upcoming models in Codex across harness changes (e.g., connectors, CUA, memory/compaction, and reasoning calibration); iterated toward a secure, reliable eval environment, stabilizing runs and safeguarding agent behavior; helped incubate an agent-driven evaluation system for R&D.

improved coding agent reliability with better context and failure analysis; explored training lower-cost subagents with multi-agent RL.

worked on early-stage LLM development behind initial Apple Intelligence launches, including Writing Tools and Reply Suggestions.

sequential recommendation and retrieval; AutoML