Kunal Jain
Hi! I am a PhD student at Georgia Institute of Technology under Prof. Divya Mahajan. My research interests lie at the intersection of Machine Learning and Systems.
Previously, I was a Research Fellow at Microsoft Research, India (M365 Research Group) with Dr. Anjaly Parayil and Dr. Ankur Mallick.
My work at Microsoft has focused on improving data center efficiency and serving low latency ML inference workloads. We are looking at ways to improve the serving of LLM workloads on a datacenter level, with multiple models and hardwares available and millions of requests coming in each hour.
Prior to Microsoft, I completed my Bachelors + Masters (by Research) from IIIT Hyderabad, where I was a Research Assistant at the Computer Systems Group. During my time at IIIT, I worked on building a scalable and distributed architecture for large scale camera networks, focused on a city to country wide scale. I also worked on Bayesian Optimization and how the algorithm can be adapted to various experimental settings.
Please feel free to reach out to me if you want to discuss any ideas or problems!
News
- Sep 2025: Our work on auto-scalers for LLM datacenters, SageServe, got acceptede to SIGMETRICS 2026! This work was part of my RFship at Microsoft. Link to the arXiv version of the paper.
- Aug 2025: Started my PhD at GeorgiaTech!
- Feb 2025: Our work on load balancer for LLM datacenters, Intelligent Router, got accepted to EuroMLSys 2025 Workshop (co-located with EuroSys and ASPLOS 2025)! Link to the arXiv version of the paper.
- Feb 2025: Our work on proactive scaling for LLM datacenters, SageServe, is available on arXiv: Link
