Avinash Kumar

prof_pic.jpg

Office: EER 5.884B | Email: avinkumar@utexas.edu

I am a PhD student in Electrical and Computer Engineering at The University of Texas at Austin advised by Prof. Poulami Das.

My research focuses on building efficient systems for ML, with a current emphasis on Large Language Models (LLMs). My work is supported by the Amazon AI PhD Fellowship and the Professor Edith Clarke Fellowship.

Before starting my graduate studies, I obtained my MS degree from UT Austin and my Bachelors degree from NITK Surathkal. I also worked for three years at NVIDIA as a Senior GPU Architect.

News

June 2026 I will be in Seattle at Amazon Science working on Reasoning LLMs for the upcoming summer!
May 2026 Presenting our work HELIOS at MLSys 2026

selected publications

  1. MLSys
    HELIOS: Adaptive Model And Early-Exit Selection for Efficient LLM Inference Serving
    Avinash Kumar, Shashank Nag, Jason Clemons, and 2 more authors
    Proceedings of Machine Learning and Systems, 2026
  2. ICML
    Quarch: A benchmark for evaluating llm reasoning in computer architecture
    Shvetank Prakash, Andrew Cheng, Arya Tschand, and 24 more authors
    In Forty-third International Conference on Machine Learning, 2026
  3. ICML
    Dialogue Without Limits: Constant-Sized KV Caches for Extended Response in LLMs
    Ravi Ghadia, Avinash Kumar, Gaurav Jain, and 2 more authors
    In Forty-second International Conference on Machine Learning, 2025