I'm a Graduate Student at the University of Wisconsin-Madison in Department of Electrical & Computer Engineering Department.
I'm currently pursuing my summer'22 internship at NVIDIA as a Systems Architect Intern with my work focusing on Reliability, Availability and Serviceability (RAS) Architectures for NVIDIA's latest Grace CPUs.
Previously, I did a co-op at Advanced Micro Devices Inc., in the Radeon Technology SoC Architecture Group. My work focused on designing simulator infrastructure and performance modeling of AMD's Instinct™ accelerators and Radeon™ RX Graphics Cards to propel HPC & AI workload analysis.
I'm a Research Assistant at UW STACS lab, under Professor Joshua San Miguel. My research focuses on building low-power computer systems using approximate computing. I'm currently working on designing a tightly-coupled hardware accelerator that offloads the predicate computation onto a reconfigurable fabric for Hard-to-Predict (H2P) branches.
I was recently awarded the DAC Young Researcher Fellowship for the 58th Design Automation Conference 2021, San Francisco. I will be a Graduate Teaching Assistant in ECE353: Introduction to Microprocessor Systems @ UW Madison!
I completed my Bachelor's from BITS Pilani, Pilani Campus, 2017 in Electrical & Electronics Engineering with a Bachelor's Thesis on Approximate Computing under Professor Akash Kumar, Chair for Processor Design, TU Dresden, Germany. Post my bachelor's, I worked with Qualcomm Snapdragon Systems team for 3.5 years on multiprocessor concurrency and system coherency; being directly involved in the tape-out of 12 Snapdragon chipsets.
A brief history of my previous work.
DAC'21 Poster Presentation
Dynamic Predication for Hard-to-Predict Branches Poster, presented in 58th Design & Automation Conference, San Francisco.
Bachelor's Thesis Presentation
Impact of approximate adders on QRS Peak detection algorithm.
Relevant Courses Completed
Advanced Computer Architecture
High Performance Computing with CUDA & OpenMP
Embedded Systems Design
Digital System Design