News

I will be serving on the Program Committee for APNet 2026.

I'm thrilled to share that I successfully defended my Ph.D. in Computer Science at Purdue University. This marks the end of an incredible journey and the beginning of an exciting new chapter.

I am honored to receive the Broadcom Research Award in recognition of my contributions to systems and networking research for large-scale AI.

Our paper, "Reimagining RDMA Through the Lens of ML," was accepted to IEEE Computer Architecture Letters (CAL), presenting a new perspective on RDMA architecture optimized for ML workloads.

I'm excited to share that I will be interning at Microsoft Research during the summer of 2025, where I will be working on Cloud and Edge Actions.

Excited to announce the launch of the OptiReduce website! Visit optireduce.github.io to learn more and try it out yourself.

OptiReduce introduces a resilient AllReduce framework that leverages bounded-loss reliability to optimize distributed deep learning, ensuring tail-optimal and predictable performance even in challenging environments. I will be presenting it at NSDI 2025 in Philadelphia, PA.

I've been selected to join the NSDI 2025 pre-review task force. It's a great honor and opportunity to contribute to the community!

I'm incredibly excited and grateful to have received the Google Research Scholar Award for my work on Tail-Optimal Collective Communication for Distributed Training.

I will present Ultima at IBM's AI Compute Symposium in November 2023, in Yorkshire, NY. This presentation will highlight how Ultima can optimize distributed AI workloads in real-world environments.

I will be presenting a live demo of Ultima at the FABRIC Knit 7 Workshop on September 27, 2023. I'm looking forward to showcasing Ultima and receiving valuable feedback.

I will be joining Hewlett Packard Enterprise (HPE) as a Research Associate to work on profiling and optimizing workload distribution in large-scale ML systems.

Ultima is a groundbreaking work that challenges the long-standing notion of 100% reliability in AI systems, and I will be presenting it at OSDI in Carlsbad, CA, in July 2022.

I'm excited to share that I will be interning at NVIDIA during the summer of 2022, where I will be working on optimizing network interconnects in HPC environments to enhance ML training performance.

FuzzUSB is a tool developed to detect vulnerabilities in USB gadget devices using novel methods. It will be presented at the IEEE Security & Privacy (S&P) conference in San Francisco, CA, in May 2022.

I will be presenting my work on "Constructing the Face of Network Data" at the SIGCOMM 2021 Poster session. This research focuses on novel methods of network traffic analysis and security.

Excited to start my Ph.D. journey at Purdue University in the Department of Computer Science, where I'm focusing on systems, networks, and AI/ML infrastructure research!