I am a Deep Learning Performance Architect at NVIDIA. My work focuses on performance modeling and inference/training optimizations.
Prior to NVIDIA, I was pursuing a PhD under Prof. Kaushik Roy in the School of Electrical and Computer Engineering at Purdue University. My research thesis is on compute-in-memory based accelerators for ML inference, spanning aspects from circuits to systems. I also worked as a Digital Design Engineer at Texas Instruments India for a year before my PhD.
We do not grow absolutely, chronologically. We grow sometimes in one dimension, and not in another, unevenly. We grow partially. We are relative. We are mature in one realm, childish in another. —Anais Nin
Beyond research, I like listening to music, dancing, solving NYT puzzles and finding guinea pigs for my cooking experiments. I also love playing badminton and have won several awards at intra- and inter- college tournaments during my undergrad at Indian Institute of Technology, Roorkee.
news
| May, 2026 | Served in the Technical Program Committee of ISLPED. |
|---|---|
| Feb, 2026 | Panel Discussion on AI Hardware. |
| Nov, 2025 | Joined NVIDIA as Senior DL Performance Architect. |
| Oct, 2025 | Defended PhD Thesis. |
| Aug, 2025 | Presented MemRaptor at ISLPED in Reykjavík, Iceland. |
| May, 2025 | Presented Dataflow-centric perspective on using CIM for matrix multiplication at ISPASS in Ghent, Belgium. |
| Mar, 2025 | Our work on hardware-software co-design for accelerating transformer inference (HASTILY) got accepted at TCAS-AI. |
| Jan, 2025 | Co-instructing the course on AI Hardware with Prof. Kaushik Roy in Spring 2025. |
| Jun, 2024 | Presented my work on “What, When, Where to Compute-in-Memory” in Design Automation Conference (DAC) as a Young Fellow. |
| Apr, 2024 | Represented Purdue at the CHIPS Summit in Washington DC. |
| Aug, 2023 | Attended MLSys Rising Stars Workshop organized by ML Commons at Google headquarters, Sunnyvale. |
| Aug, 2023 | Grateful to co-host MLSys podcast with Student@systems featuring Prof. Gennady Pekhimenko and Prof. Cheng Tan. |
| Sep, 2022 | Our review paper on Compute-in-Memory Technologies and Architectures for Deep Learning Workloads is available to read from TVLSI. |
| Aug, 2022 | Filed patent with HPL on near memory accelerator to handle concurrent inserts-lookups during hashing. |
| May, 2022 | Research internship with Heterogenous Platforms Lab (HPL), Intel at Hillsboro, Oregon. |
| Jul, 2021 | Published my first work on sparsity-aware SOT-MTJ based crossbars device-circuit co-design at ISLPED. |
| Jun, 2021 | Won Qualcomm Innovation Fellowship, North America with Mustafa Ali. |
| May, 2021 | Industry internship with Memory Subsystem team at Nvidia. |
| Dec, 2020 | Proposal for QIF selected for final round. |
| Mar, 2020 | End of face-to-face classes. Covid era started. |
| Aug, 2019 | Started Graduate Student Life at Purdue. |