Experience

Senior Privacy Engineer (2024 - Present)
Amazon
I lead technical initiatives to mitigate global regulatory risks in Amazon's retail infrastructure. In this role, I defined Privacy-by-Design strategies for large global business units, and deployed GenAI-powered privacy tools, including interactive chatbots and multi-agent systems, to automate privacy compliance and data lineage reporting.
Privacy Engineering Manager (2022 - 2024)
Snap Inc.
I directed a team of engineers productionizing privacy technologies across Snap's product portfolio, including Snapchat, Bitmoji, and Spectacles. I designed, measured and improved privacy engineering metrics by as much as 400% to improve product iteration speeds.
Privacy Engineer (2018 - 2022)
Snap Inc.
I designed and developed privacy technologies impacting the user data lifecycle, including Snap's technical response to Apple's ATT framework. My work involved architecting production-grade solutions using differential privacy and cryptography, resulting in several granted patents for privacy-preserving attribution and recommendation systems.

Education

Carnegie Mellon University, Pittsburgh, USA (2012 - 2018)
PhD, Electrical and Computer Engineering
Thesis: Fairness and Privacy Violations in Black-Box Personalization Systems: Detection and Defenses
Advisor: Anupam Datta
Indian Institute of Technology, Kharagpur, India (2008 - 2012)
B.Tech. (Hons), Computer Science and Engineering
Thesis: Towards a Faster Fully Homomorphic Encryption Scheme
Advisor: Debdeep Mukhopadhyay

Technicolor, Los Altos (Summer 2015)
Mentors: Nadia Fawaz and Marc Joye
Developed state-of-the-art techniques for private data aggregation using cryptographic and information-theoretic techniques.
Microsoft Research, Bangalore (Summer 2014)
Mentor: Saikat Guha
Identified data-flows in big data applications from logs without access to source scripts; implemented using C# and Scope on the Cosmos platform.
Max Planck Institute for Software Systems, Saarbrücken (Summer 2012)
Mentors: Michael Backes and Aniket Kate
Explored algorithms for Oblivious RAM (ORAM) to improve complexity and prevent access pattern leakage during remote data access.
Max Planck Institute for Software Systems, Saarbrücken (Summer 2011)
Mentors: Michael Backes and Aniket Kate
Devised a new Asynchronous Verifiable Secret Sharing (AVSS) protocol with O(kn²) communication complexity, an order of magnitude improvement over existing results.