OptiNIC: A Resilient and Tail-Optimal RDMA NIC for Distributed ML Workloads
Ertza Warraich, Ali Imran, Annus Zulfiqar, Shay Vargaftik, Sonia Fahmy, Muhammad Shahbaz
Under submission
I am an Academic Affiliate in the Department of Computer Science and Engineering at the University of Michigan, Ann Arbor. I completed my Ph.D. in Computer Science at Purdue University, where I worked in the PurNET and NextGArch labs under the supervision of Prof. Muhammad Shahbaz and Prof. Sonia Fahmy.
My research examines how communication mechanisms and system design choices shape the performance, reliability, and security of large-scale workloads, including modern machine learning pipelines. I work broadly in systems and networking, with interests spanning secure system architectures, distributed system design, communication efficiency, and resilience under real-world deployment constraints.

Ertza Warraich, Ali Imran, Annus Zulfiqar, Shay Vargaftik, Sonia Fahmy, Muhammad Shahbaz
Under submission
Ertza Warraich, Ali Imran, Annus Zulfiqar, Shay Vargaftik, Sonia Fahmy, Muhammad Shahbaz
In IEEE Computer Architecture Letters (IEEE CAL 2025)
Virtual, Online. December 2025
Ertza Warraich, Omer Shabtai, Yonatan Piasetzky, Shay Vargaftik, Matty Kadosh, Lalith Suresh, Muhammad Shahbaz
In Proceedings of the 22nd USENIX Symposium on Networked Systems Design and Implementation (NSDI 2025)
Philadelphia, PA. April 2025
Kyungtae Kim, Taegyu Kim, Ertza Warraich, Byoungyoung Lee, Kevin Butler, Antonio Bianchi, Dave (Jing) Tian
In Proceedings of the 43rd IEEE Symposium on Security and Privacy (IEEE S&P 2022)
San Francisco, CA. May 2022
03.
I'm always open to discussing research collaborations, interesting problems in systems and AI infrastructure, or just saying hello.
Design inspired by Brittany Chiang & Devon Gifford