Launched 14+ major features for Capacity (GPU/CPU) Management that achieved $100M+ in savings, while shifting to own Notification Settings, then Model Management designs mid-half after reorg, supporting 3 designers to onboard to prior areas.
Led end-to-end product design across internal CPU/GPU management UI to improve AI Infra capacity governance, efficiency, and ML developer productivity, growing the 0-1 tooling to a canonical surface to manage operational AI quotas supporting 2.2k active users and helped justify +3 engineering headcount to scale the product. 660 engineering hours/year saved via redesigned IA for faster navigation, 50% fewer user support posts, and major adoption lifts that grew WAU by 600%.
