Research Scholar Fellow
Research Scholar Fellow | UC Berkeley Sky Computing Lab
I am a wireless and networked systems researcher currently serving as a Research Scholar Fellow at UC Berkeley's Sky Computing Lab, where I contribute to the SkyPilot project with Professor Ion Stoica. I have been fortunate to work with exceptional researchers throughout my journey: Prof. Tamer Nadeem at Virginia Commonwealth University during my PhD, Prof. Sylvia Ratnasamy and Prof. Scott Shenker at UC Berkeley's NetSys Lab, Dr. Juan A. Fraire and Prof. Herve Rivano at INRIA Lyon, Prof. Franck Rousseau at LIG Lab (Universitรฉ Grenoble Alpes), Prof. Gil Zussman at Columbia University's Wireless & Networking Lab, and industry leaders at Qualcomm, Skylark Wireless LLC, and Bell Labs.
My work lies at the intersection of networked systems, wireless communication, distributed learning, and formal verification. My research focuses on developing techniques that enable systems to anticipate future conditions and adapt proactively rather than reactively while maintaining mathematical guarantees about performance, security, and reliability. I introduced the principle of predictive adaptation with formal guarantees: exploiting predictability in orbital mechanics, traffic patterns, and wireless dynamics to transform infrastructure from reactive to proactive systems that prevent problems before they occur. My current work focuses on developing techniques and tools that enable engineers to build adaptive systems with well-understood safety and performance properties, spanning predictive datacenter networks, space-terrestrial integration, privacy preserving and safe learning-based protocols. These techniques are deployed in production at companies including Qualcomm and Skylark Wireless, with frameworks adopted by cloud providers and protocols evaluated by satellite operators for LEO constellations. This work draws on insights and techniques from wireless communication, distributed systems, reinforcement learning, formal verification, and orbital mechanics.
My research bridges theory with practice through large-scale deployments: MU-MIMO schedulers serving millions of 5G users at Skylark Wireless (23% spectral efficiency improvement), Wi-Fi 7 protocols integrated into Qualcomm's commercial chipsets, carbon-aware schedulers adopted by major cloud providers, and open-source tools (SimSAT, SATMetrics) accelerating satellite communications research globally.
I am actively updating this site. Thanks for checking back soon.
Our paper "CacheCatalyst" has been accepted to NSDI'25.
Our paper "Balancify" has won Best Paper and Best Contribution Award at ACM CoNEXT'25.
I co-chaired ACM CoNEXT Student Workshop, and gave a talk on "AI for Networks and Networks for AI".
Our three papers were presented in MobiCom'25.
Our latest research on datacenter microbursts has been accepted to ACM SIGMETRICS 2025.
Presented our work on carbon-aware scheduling at the Sustainable Computing Workshop.
Received Best Paper Award for "Role of Machine Learning in Satellite IoT".
Received research grant for sustainable AI infrastructure development.
New partnership with major cloud provider for carbon-aware scheduling implementation.
Released SimSAT and SATMetrics tools for the satellite communications research community.
Framework for carbon-aware computing and sustainable AI systems. Developing AI infrastructure that reduces global ICT carbon emissions while maintaining performance and reliability across the edge-to-datacenter continuum.
Research in on-device ML including edge computing, privacy-preserving technologies, and hardware accelerators. Focus on algorithms that maintain privacy while optimizing performance at the network edge.
Advanced research in satellite IoT systems, direct-to-device communications, and space-based network architectures. Developing machine learning approaches for next-generation satellite networks.
Interested in collaboration, research discussions, or musical exchanges? Feel free to reach out!