Hardik joined Bigstream in 2018 while simultaneously finishing up his PhD at Georgia Tech. He leads the hardware acceleration team at Bigstream and is responsible for analyzing big data/machine learning/deep learning workloads, finding opportunities and architecting solutions for hardware acceleration, and designing the hardware-software interfaces to enable seamless integration with Bigstream’s Hyper Acceleration technology.
During his PhD, Hardik researched domain specific hardware architectures for accelerating machine learning, especially deep neural networks. He led the development of the very first open-source FPGA-based hardware acceleration stack for deep neural networks at Georgia Tech. His research led to several publications in top-tier computer architecture conferences. His research was recognized by the prestigious Qualcomm Innovation Fellowship (QInF) in 2018.
Upon finishing his undergrad education at IIT Guwahati, Hardik started his career at Qualcomm, India in 2013, where he worked on FPGA-based emulation of Qualcomm chips.