Built an off-policy RL next-best-product optimizer on customer time-series data (TensorFlow, distributed GPU), achieving 85% top-3 DD recall; deployed via Vertex AI pipelines; patent filing in process
Developed Transformer-based propensity models, improving top-1 DD recall by ~30% (OOT); optimized audience selection via Lagrangian post-scoring, driving ~12% incremental revenue lift
Engineered sparse time-series feature pipelines across 50M+ households (across 52 weeks horizon), reducing training time by ~35%
Designed and executed A/B tests, delivering +10 bps conversion and +7 bps CTR lift
Graduate Student Research Assistant
University of Michigan
Ann Arbor, Michigan
01.2024 - 04.2025
Designed and implemented large-scale data processing pipelines using SQL databases for residential property datasets (100M+ records), enabling scalable record linkage and wealth analysis
Developed probabilistic record linkage models using Expectation-Maximization (EM) and text similarity algorithms (Jaro-Winkler), improving matching accuracy upto 84% across datasets
Leveraged high-performance computing (HPC) environments to parallelize data processing workflows, significantly improving runtime efficiency by 24% for large-scale entity resolution tasks
Data Analyst Intern
Tesla
Fremont, California
05.2023 - 08.2023
Developed a MILP based optimization model using Python and Gurobi, integrated with BI tools, to recommend optimal carrier allocation, cutting manual work by 80% and saving $20M annually.
Built demand forecasting components and data transformation ETL pipelines using statistical modeling and R, improving inventory planning and imbalances detection accuracy
Business Intelligence Engineer
Amazon
India
06.2020 - 06.2022
Developed Bayesian PED models (~7.5% MAPE) to optimize pricing decisions, enabling advanced shipping carrier selection with ~8% sales attrition on higher ASP segments
Built predictive models for seller affiliate payout estimation using historical data and business heuristics, enabling accurate compensation payouts forecasting for 4.5k affiliate partners
Implemented automated marketing solutions using Redshift and AWS Lambda, resulting in a 38% increase in seller registration rates: refined strategies through comprehensive A/B testing that elevated campaign ROI by 15%
Developed automated reporting workflows in analytics solutions Selenium and Python, reducing manual reporting effort and improving turnaround time for business insights by ~2 days
Associate Data Scientist
United Airlines
06.2019 - 05.2020
Developed aircraft maintenance risk prediction models deployed in Palantir Foundry, reducing workload deferrals by ~55% and improving operational efficiency
Optimized PySpark-based data pipelines for large-scale operational datasets, reducing compute time by ~80% and improving system uptime by ~20%
Education
MS - Computer Science & Engineering
University of Michigan
Ann Arbor, MI
05-2025
MS - Industrial & Operations Engineering
University of Michigan
Ann Arbor, MI
05-2025
B.Tech - Aerospace Engineering
Punjab Engineering College
India
05-2019
Skills
Programming: Python, SQL, C, R
Frameworks: PyTorch, TensorFlow
Cloud: GCP, AWS, Palantir
MLOps: Docker, Kubernetes, ADK
Publications
Numerical Study of Deflagration to Detonation Transition using OpenFOAM (C++), IEEE Aerospace Conference (2020)
Numerical Study of Deflagration to Detonation Transition using OpenFOAM (C++), Springer (2019)