Harshul Gupta

Harshul Gupta Email and Phone Number

Software Engineer at Amazon Prime Air (Perception) @ Amazon
Harshul Gupta's Location
Seattle, Washington, United States, United States
Harshul Gupta's Contact Details

Harshul Gupta personal email

n/a
About Harshul Gupta

I am a software engineer working on autonomous drone delivery within the perception team at Amazon Prime Air. In 2020, I finished my graduate studies at the University of California San Diego with a Master's in Computer Engineering. My key area of interest lies in the field of Deep Learning and I have worked on several problems pertaining to GANs, Engagement Detection, and Text to Image Generation. Github: https://github.com/harshulgupta2

Harshul Gupta's Current Company Details
Amazon

Amazon

View
Software Engineer at Amazon Prime Air (Perception)
Harshul Gupta Work Experience Details
  • Amazon
    Software Engineer
    Amazon Apr 2020 - Present
    Seattle, Wa, Us
  • Outward, Inc.
    Computer Vision Research Intern
    Outward, Inc. Jun 2019 - Dec 2019
    San Jose, Ca, Us
  • University Of California San Diego
    Graduate Teaching Assistant
    University Of California San Diego Apr 2019 - Jun 2019
    La Jolla, Ca, Us
    Teaching Assistant for ECE 228 - Machine Learning for Physical Applications under the guidance of Professor Peter Gerstoft.
  • University Of California San Diego
    Graduate Teaching Assistant
    University Of California San Diego Jan 2019 - Mar 2019
    La Jolla, Ca, Us
    Teaching Assistant for CSE 148 - Advanced Processor Architecture Design under the guidance of Professor Dean M. Tullsen.
  • International Institute Of Information Technology – Bangalore
    Machine Learning Research Intern
    International Institute Of Information Technology – Bangalore Apr 2018 - Jul 2018
    Bengaluru, Karnataka, In
    ● Part of Multimodal Perception Lab headed by Professor Dinesh Babu Jayagopi● Designed a platform based on Python for a Socially Interactive Virtual Language Tutor for teaching the French language to the end users. Major components of the platform were◦ Facial Recognition Module◦ Engagement Detection Module◦ MARC Virtual Agent and Language Translation Module● Built a face detector using YOLOv2 [computed anchor-boxes using k means] and an SVM based engagement detector using OpenFace [F1 score: 0.82; ZeroMQ to get OpenFace features].● Modeled a sentiment recognition module using Keras to assess user’s intent to engage with the system; used it along with Pocketsphinx ASR to model agent’s expressions [positive/negative] by performing sentiment-analysis on the text.
  • Western Digital
    Senior Engineer, R&D Engineering (Memory Technology )
    Western Digital Jul 2017 - Apr 2018
    San Jose, Ca, Us
    ● Designed C based Firmware APIs to statistically characterize the performance and endurance of NAND Flash across varied fabrication lots using an ARC processor. Worked with GoLogic and JTAG debugger to resolve the reported firmware issues.● Minimized the load on BCH and LDPC unit of ASIC by developing novel algorithms to detect optimal read voltages.● Improved the existing write abort detection model to avoid miss-detection of an unforeseen power droop during program operation. Presented the algorithm at an internal conference of Western Digital Corporation.● Created a tool using Python for generating the failing command sequence from GoLogic Waveform to accelerate the memory debug process.
  • Western Digital
    Engineer, R&D Engineering (Memory Technology)
    Western Digital Jun 2015 - Jun 2017
    San Jose, Ca, Us
    Co-Inventor, Patent (USPTO) Dynamic Read Table Generation◦ Utilized the soft-bit reads to find optimal read voltages dynamically for an MLC NAND Flash which compensated for the need of stronger ECC by reducing the overall read-errors and helped in keeping the production cost low.◦ SanDisk Innovation Award 2016 recipient.
  • Broadcom
    Audio Firmware Division ( Internship )
    Broadcom Jan 2015 - Jun 2015
    Palo Alto, California, Us
    ● Optimized the fixed point C models of audio algorithms to get the optimal performance on a 64-bit dual SIMD, RISC architecture based DSP processor for a set-top box (STB) chip. Considering the importance of MAC in DSP, supporting faster MAC operations was essential● Efficiently employed loop unrolling along with vectorization for the most cycle intensive functions to reduce the overall execution cycles. Worked primarily on PCM Wave, MPEG-4 ALS, and Dolby Multistream Decoder

Harshul Gupta Skills

C C++ Flash Memory Perl/python Verilog Matlab Microsoft Excel Latex Guitar Playing Microsoft Powerpoint Data Analysis C# Tensorflow Keras Jtag Pytorch Cuda Deep Learning Computer Vision Opencv

Harshul Gupta Education Details

  • Uc San Diego
    Uc San Diego
    Computer Engineering
  • Birla Institute Of Technology And Science, Pilani
    Birla Institute Of Technology And Science, Pilani
    Electronics And Instrumentation

Frequently Asked Questions about Harshul Gupta

What company does Harshul Gupta work for?

Harshul Gupta works for Amazon

What is Harshul Gupta's role at the current company?

Harshul Gupta's current role is Software Engineer at Amazon Prime Air (Perception).

What is Harshul Gupta's email address?

Harshul Gupta's email address is ha****@****isk.com

What schools did Harshul Gupta attend?

Harshul Gupta attended Uc San Diego, Birla Institute Of Technology And Science, Pilani.

What skills is Harshul Gupta known for?

Harshul Gupta has skills like C, C++, Flash Memory, Perl/python, Verilog, Matlab, Microsoft Excel, Latex, Guitar Playing, Microsoft Powerpoint, Data Analysis, C#.

Free Chrome Extension

Find emails, phones & company data instantly

Find verified emails from LinkedIn profiles
Get direct phone numbers & mobile contacts
Access company data & employee information
Works directly on LinkedIn - no copy/paste needed
Get Chrome Extension - Free

Aero Online

Your AI prospecting assistant

Download 750 million emails and 100 million phone numbers

Access emails and phone numbers of over 750 million business users. Instantly download verified profiles using 20+ filters, including location, job title, company, function, and industry.