Hardik Sharma

Hardik Sharma

Director of Hardware Engineering, Bigstream

Georgia Institute of Technology (GaTech)

Biography

Hardik Sharma is the Director of Hardware Engineering at Bigstream (https://bigstream.co) where he 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.

Hardik received his PhD from the department of Electrical and Computer Engineering at Georgia Tech in 2019, supervised by Prof. Hadi Esmaeilzadeh. His research interests are domain specific hardware architectures for accelerating machine learning, with a focus on deep learning. He led the development of the first open-source FPGA-based hardware acceleration stack for DNNs at Georgia Tech (http://dnnweaver.org). His research has been published in several top-tier conferences and journals. His work was recognized by the prestigious Qualcomm Innovation Fellowship (QInF, 2018).

Prior to joining Georgia Tech, Hardik worked in Qualcomm India as an FPGA Engineer upon finishing his undergrad education at the Indian Institute of Technology (IIT) in Guwahati, India in 2013.

Interests

  • Deep Learning
  • Computer Architecture
  • Domain-specific Hardware Architectures

Education

  • PhD in Electrical and Computer Engineering, 2019

    Georgia Institute of Technology

  • MS in Electrical and Computer Engineering, 2015

    Georgia Institute of Technology

  • B.Tech in ECE, 2013

    Indian Institute of Technology, Guwahati

Publications

Mixed-Signal Charge-Domain Acceleration of Deep Neural networks through Interleaved Bit-Partitioned Arithmetic

Planaria: Dynamic Architecture Fission for Spatial Multi-Tenant Acceleration of Deep Neural Networks

Bit-Parallel Vector Composability for Neural Acceleration

From Tensors to FPGAs: Accelerating Deep Learning

Bit fusion: Bit-level dynamically composable architecture for accelerating deep neural network

Experience

 
 
 
 
 

Director of Hardware Engineering

Bigstream

Jul 2019 – Present Mountain View, CA
 
 
 
 
 

Lead Product Engineer, Deep Learning Solution

Bigstream

Aug 2018 – Jul 2019 Mountain View, CA
 
 
 
 
 

Research Intern, R&D Machine Learning for IoT (ML-IoT) group

Arm, Inc

May 2017 – Aug 2017 San Jose, CA
 
 
 
 
 

Research Intern, Media Processing Group (GPU Architecture) and R&D

Arm, Inc

May 2016 – Aug 2016 San Jose, CA
 
 
 
 
 

Engineering Intern, SoC Design

Qualcomm

May 2015 – Aug 2015 San Diego, CA
 
 
 
 
 

Associate Engineer, Emulation Team (Full-time)

Qualcomm

Jul 2013 – May 2014 Bangalore, India
 
 
 
 
 

Engineering Intern, Emulation Team

Qualcomm

May 2012 – Jul 2012 Bangalore, India
 
 
 
 
 

Engineering Intern, Space Applications Centre (SAC)

Indian Space Research Organization (ISRO)

May 2011 – Jul 2011

Contact

  • hardik[at]bigstream[dot]co