Hello! I am Shreya, a Software Engineer at Meta I am working for Facebook Search Integrity, building systems that curb the prevalence of very harmful content on the site. Before this, I was a master's student at School of Computer Science at Carnegie Mellon University. During my master's, I completed a summer internship at Meta as a Backend Software Engineer. At Meta, I was part of the storage team building second-generation of blob storage for providing our clients with more flexibility in terms of assigning a storage policy. In particular, I built a C++ service for the automated generation and management of storage volumes.

My research interests primarily focus on secure systems that allow multiple parties to perform decentralised computation without much overhead. I'm currently working on private information retrieval with Prof Thomas Schneider.

In the summer prior to my master's, I worked as a software engineer for the Bing UX Team at Microsoft India. My tasks at Bing mainly included - bug fixing and incorporating feedback to the recently launched TravelHub. For my Bachelor’s Thesis, I collaborated with ENCRYPTO Group at TU Darmstadt to build a framework for secure multi-party computation (MPC) called MOTION. We've used various technologies and design decisions to build MOTION in a user-friendly and extensible manner, and I hope that it promotes adoption of MPC protocols in practice.

In 2019, I interned at NTU Singapore to work on collaborative Deep Learning (DL). My research involved designing efficient protocols for federated learning on DL models while strengthening the security guarantees involved. Previously, I also worked with Dr Carmit Hazay at Bar Ilan University, Israel to come up with mathematical constructs that make way for simpler protocols in MPC.

Publications

• Shreya Sharma, Chaoping Xing, Yang Liu, Yan Kang. "Secure and Efficient Federated Transfer Learning". In 2019 IEEE International Conference on Big Data, Los Angeles, CA, USA (pp. 2569-2576). IEEE.
• Shreya Sharma, Chaoping Xing, Yang Liu. "Privacy-preserving deep learning with SPDZ" The AAAI Workshop on Privacy-Preserving Artificial Intelligence 2020, New York, NY, USA.

Professional Activity

• Program Committee Member for:
• ACM CCS - ARTMAN 2025
• EMNLP - Perspectivist Approaches to NLP 2025
• NeurIPS 2025 Workshop on Continual and Compatible Foundation Model Updates (CCFM)
• 24th IEEE International Conference on Trust, Security and Privacy in Computing and Communication
• International Generative AI and Computational Language Modelling Conference (GACLM) 2025
• Big Data Analytics & Applications (BDAA) 2025 Conference
• IEEE Bigdata - Applied Artificial Intelligence for Public Safety and Security Workshop on Big Data (AI-PublicSecurity-2025)
• International Workshop on Big Data & AI Tools, Methods, and Use Cases for Innovative Scientific Discovery (BTSD) 2025
• Computer Networks Journal
• Member IEEE and Trust & Safety Professional Association.
• Teaching Assistant Foundations of Blockchains and Intro to Data Science

Open-Source

• Secure ML: repository implements Secure Linear Regression in the Semi-Honest Two-Party Setting as laid out in the SecureML paper published in IEEE S&P
• MOTION: This is a framework for Mixed-Protocol Multi-Party Computation in a Semi-Honest Setting as laid out in the MOTION paper published in ACM TOPS

Awards and Achievements

• Awarded the Google - Grace Hopper Celebration (GHC) Scholarship to attend GHC 2020.
• Awarded the DAAD-WISE scholarship 2020 by German Government for research at TU Darmstadt.
• Presented my work on secure Deep Learning at AAAI 2020.
• Selected for Microsoft - Codess Summer Mentorship Program 2019.
• Awarded the SPARK 2019 Scholarship by IIT Roorkee.