About Me

Hi, I'm Ushasi ๐Ÿ‘‹ a fifth-year Ph.D. student in Electrical & Computer Engineering at UC San Diego, building the next generation of AI-native wireless networks.

I work with Prof. Dinesh Bharadia in the WCSNG lab. Before my Ph.D., I earned my MS at UCLA and B.Tech at NIT Durgapur, and worked at Qualcomm on 5G NR protocol optimizations for extended reality (XR) applications.

๐Ÿ’ผ On the job market: Looking for full-time research roles. See my CV and project website, or contact me at ughosh@ucsd.edu.
Networked Systems AI/ML Optimization Edge-native 5G & Open RAN Application QoE Physical AI Satellite Networks (NTN)

Research Overview


My long-term vision is to build user-centric networked systems that connect application-level experience (QoE) with network control. Despite rich cross-layer telemetry, todayโ€™s networks remain largely unaware of how their decisions affect applications. This gap is particularly important for latency-sensitive workloads such as streaming, XR, and interactive AI, and even more so for robotics, where network delays can directly affect task success. My research explores how network behavior and control decisions shape application performance, with the goal of moving beyond traditional network metrics toward optimizing user experience and task outcomes.

๐Ÿ† My current project, mmsRAN: Enabling Physical AI with the Cellular Edge, is funded by the Qualcomm Innovation Fellowship 2026.
๐Ÿš€ Explore my flagship project, EdgeRIC: real-time intelligent optimization and control for NextG cellular networks.

Selected Talks


EdgeRIC tutorial - Real time AI-in-the-loop for Cellular Networks

srsRAN Workshop 2024 October 2024 Arlington, Washington D.C
Workshop Tutorial
Watch

Hands-on tutorial session: EdgeRIC

Mississippi State University August 2024 Starkville, MS
Workshop Tutorial
Watch

EdgeRIC: Empowering Real-time Intelligent Optimization and Control in NextG Cellular Networks

NSDI 2024 April 2024 Santa Clara, California
Conference Proceedings Talk
Watch