Jack of all trades at an early-stage startup working on bleeding edge ML problems. Worked across the stack from core cloud infra, to python backend APIs, to frontend React UI, improving productivity and achieving key company goals.

Lead migration of full storage stack & backend libraries from custom solution to Postgres + SQLAlchemy, improving latencies by 2-3 orders of magnitude & fixing longstanding UX & system issues.

Created tooling on top of storage library created for postgres migration to unlock faster research workflows for evaluating ML model performance, while seamlessly integrating with standard ML tooling (e.g. Jupyter notebooks).

Lead team maintaining critical Billing UX ETL pipelines, working with management to solve critical reliability & latency issues, and co-lead design of next-generation billing pipeline.

Solved longstanding, crippling performance problems. Fixed frequent outages: <90% uptime from >100x input scale spikes. Left team with >99.5% reliability and a quiet pager.

Proud mentor: hosted an intern who later converted into a full time engineer, directed several junior engineers on the team, lead an Effective Java reading group.