Jul 2020 – Mar 2023 · 2 yrs 8 mos · San Francisco, California, United States

• Tech Lead: Graph Mining for Ads Targeting and Modeling (> 1 year)
• Led Graph ML driven improvements for following key products with > 4% increase in Ads Revenue
• 1) Candidate Sourcing for Ads Ranking
• 2) Interest Targeting
• 3) Lookalike Expansion or Lookalikes
• Multi-Stage Ads Ranking Improvements
• Probabilistic Matching solutions for iOS 14 (App Tracking Transparency)

May 2019 – Aug 2019 · 3 mos · Greater New York City Area

• Worked on the problem of fraud detection in Credit Card Transactions.
• Obtained improved F1 scores over prior art with published US Patent US20220012741A1.
• Devised a Deep Learning approach along with experiments using ensemble models.
• Dealt with heterogeneous features, extreme class imbalance and anomalies.

Jul 2017 – Jul 2018 · 1 yr · Microsoft Research, Bangalore

• Project titled ”Unsupervised representation learning on heterogeneous graphs”.
• Devised an online algorithm and tackled issues with representation based ranking.
• Performance of online algorithm within 5% compared to that of static setting.
• Learned representation to be used for a variety of intelligent tasks in Outlook.

May 2016 – Jul 2016 · 2 mos · Bengaluru Area, India

• Worked on ”Usage-based Prototype Evaluation” for ”Mobile User Intelligence”.
• Employed language modeling techniques and topic modeling algorithms for solution.
• Developed a prototype for demonstrating the use cases before Adobe Research lab.

May 2015 – Jun 2015 · 1 mo · Udaipur Area, India

• Worked on a chat based assistant application for ordering and booking along with a recommendation module.