Sagar Jha

Experience

• 

  FANUC India Pvt Ltd

  Aug 2016 - Aug 2020

  • Sr. Deputy Engineer (Robot System Engg)

    Apr 2018 - Aug 2020

  • Deputy Engineer (Robot System Engg)

    Aug 2016 - Mar 2018
• 

  Thapar Institute of Engineering & Technology

  Sept 2021 - Dec 2022

  Deep Learning Research Intern

  Worked on a research project involving Heart Rate Estimation using Vision Transformers to predict rPPG signals from Facial videos.We propose a JTCNet for rPPG based HR estimation. JTCNet incorporates a local spatial-temporal change modeling network based on CNN with spatial and frequency domain losses, and a long-range temporal change modeling network based on Transformer.

• 

  Indian Institute of Technology, Kharagpur

  Nov 2021 - Feb 2022

  Computer Vision Intern

  Developed a CV-based system using quadcopter live feeds to detect crops and monitor intruder activity. Implemented DL models (YOLOv4/v5, SSD, MobileNet, VGG16) for crop detection, plant disease, pest, and nutrition monitoring.Created a crop dataset of local Indian crops, training models for accurate detection and health assessment

• 

  Mount Royal University

  Jun 2022 - Aug 2022

  MITACS-GRI'22 - Globalink Research Intern

  Research-based summer internship at the Mathematics and Computing dept, MRU, under Prof. Yasaman Amannejad's mentorship in the Distributed Machine Learning domain (Federated and Split Learning).We worked on the idea that can a server utilize deep learning models for training or inference without accessing raw data from clients? Split learning naturally allows for various configurations of cooperating entities to train (and infer from) machine learning models without sharing any raw data or detailed information about the model.Developed 6 different deep learning architectural algorithms using Split Learning and Federated Learning techniques. Explore the benefits of one algorithm over another using various experiments and analyses. The algorithms were implemented on Edge devices (RPi, Jetson Nano, etc.) and on huge compute servers to accommodate different configurations of computing devices and their effects on the algorithms. Show less

• 

  JPMorgan Chase & Co.

  Jan 2023 - now

  Software Engineer

  My team monitors JPM's suite of trading algorithms by generating 400K+ real-time order status and performance updates per day using a KDB+ based alerting engine. We also produce 3000+ trading signals per day via a chatbot that mines trader discussions and news using Natural Language Understanding (NLU).