Shreyas’ personal website

Ravi Shreyas Anupindi is a research engineer at Microsoft Research India, where he works with the AI Infrastructure and Extreme Classification teams. He is interested in building large-scale machine learning systems, and has previously worked on Project Forge, Microsoft Azure’s GPU infrastructure for AI workloads.

Some of the projects he’s worked on at Microsoft include:

• Transparent checkpointing and elasticity for DNN training workloads (Project Forge).
• Distributed DNN training and inference storage cache for large-scale AI workloads (Project Forge).
• QoS aware scheduling for LLM server replicas.
• Scaling large scale retrieval models to billions of parameters.
• Global routing and provisioning for LLM inference workloads.

Before joining Microsoft, he was a research assistant at the Computer Systems Lab in the Department of Computer Science and Automation, Indian Institute of Science. He worked on accelerating key-value stores using page table walkers.

He holds a B.Tech in Computer Science and Engineering from PES University Bangalore.

In addition to his professional pursuits, Shreyas is an avid reader and gamer, and you can find a list of some of the books and games he’s played on his Goodreads and Steam profiles (links in sidebar). When he’s not working, gaming or reading, he lives a pretty normal life hanging out with friends, doing mundane things and enjoying the little things in life.

You can find some of his writings here: Substack