Experience

Machine Learning Systems Engineer

Eight Sleep

  • Fully architected a model training stack to scale up data and compute — self-hosted Anyscale on AWS using aws-cdk, dockerized environments for CI, EC2, and Anyscale, and built a config-driven training paradigm using omegaconf and ray
  • Saved ~$60K/year by reducing the memory footprint of a production Kinesis pipeline through signal processing algorithm optimizations
  • Sped up dataset preparation and model evaluations ~100x by moving local workflows to AWS ECS
  • Developed observability pipelines that detected and patched failures in 10% of devices in the field
  • Built live data streamer and various data parsers to enable experiments with new sensors
aws-cdkec2ecskinesissqsdynamo-dbpostgresqls3nodepythontorchonnxrayanyscaledockerdatadog

Machine Learning Software Engineer

Microsoft

  • Developed a LLM-As-A-Judge metrics pipeline to enable shipping decisions for Microsoft 365 Copilot
  • Enabled anomaly detection along with SHAP-based model interpretability for IT admins
azure-mlazure data factorysparkxgboostlightgbmgpt-3/3.5/4python.NET

Graduate Research Assistant

Pendar Technologies

  • Conducted research on identification of chemicals using infrared imaging and Raman spectroscopy by solving a variant of the dictionary learning problem
researchdictionary learningpython

Research Engineer Intern

Avigilon, A Motorola Solutions Company

  • Prototyped, validated, and deployed camera auto-calibration for computer vision applications by transforming a geometric problem into an optimization problem
convex optimizationcamera calibrationpythonc++

Education

M.S. and B.S. in Electrical Engineering

Tufts University

  • GPA 3.9, Dean’s List all semesters
  • Member of Tau Beta Pi and IEEE-HKN