Jan 2023 – Sep 2024 · 1 yr 8 mos · Bellevue, Washington, United States

• Part of the core video recommendation team.
• Bridged the gap between academic research and industry by designing and implementing state of the art recommendation methods that led to key metric improvements.
• Focus: Generative Recommendation, Sequence to Sequence modeling, fault tolerant multi-data center recommendation design

May 2021 – Aug 2021 · 3 mos

• Research on fairness and transparency in machine learning

Jan 2021 – May 2021 · 4 mos · Seattle, Washington, United States

• Research on mechanism design and machine learning.

Aug 2020 – Dec 2020 · 4 mos

• Part of the Economics, Algorithms, and Optimization team in Core Data Science.
• Worked on designing automatic bidding strategies for ad campaigns with budgets.
• Developed a modular pipeline that first estimates the optimal budget spend plan using historical data and then uses pacing system to match the learned spend plan.
• Research paper presented at The INFORMS Annual Meeting 2021.

May 2019 – Aug 2019 · 3 mos · Menlo Park, California

• Machine learning in the ads ranking team. Worked on adding value optimization to product ads ranking.
• Developed the complete pipeline including creating data pipelines, training deep learning models, and updating the ads ranking service to include value optimization predictions.
• Introduced an optimized architecture for the model which led to over 84% reduction in model size, an overall reduction of over 80 TB of RAM usage across deployed machines. It also led to faster response time and increased prediction accuracy.

May 2016 – Jul 2016 · 2 mos · Baltimore, Maryland Area

• Research internship in the machine learning theory group in the Department of Computer Science with Dr. Raman Arora.

May 2015 – Jun 2015 · 1 mo · Dubai, United Arab Emirates