Seeking Software Engineering roles in domain of Deep learning and CV

Shubham Turakhia

adivel@r.postjobfree.com 480-***-**** linkedin.com/in/shubham-turakhia/ github.com/Shubhamturakhia SUMMARY

Currently seeking Software Engineering roles focused in domain of Deep Learning, Computer Vision and Robotics EXPERIENCE

Software Engineer Intern, Computer Vision, Sensagrate, Scottsdale, AZ May 2020 - Aug 2020

• Project 1: Developed code infrastructure for pedestrian safety applications with expectation of getting >20 FPS

• Achievement : 1. Algorithm developed using Tensorflow 2.0 framework using real-time data captured from Camera and LiDAR sensors. The code encapsulated features of multi object detection, tracking, making polylines, zones and ignore zones with easy deployment of models with minimal changes 2. Achieved 33 FPS (Yolo v3 model) with TensorRT Optimization on Jetson TX2. 3. It was tested with other models (EfficientDet D0 & D4, MobileNetSSD) and using additional methods (Federated & Transfer Learning)

• Project 2: GUI Development for creation of zones and polylines on point cloud data and images from stationary sensors

• Achievement: Created a customized GUI using Rviz and was validated on pedestrian safety project Graduate Student Researcher, Robotics and Intelligent Systems Laboratory, Mesa, AZ Aug 2019 - Dec 2020

• Masters Thesis Project : A Deep Reinforcement Learning Approach for Robotic Bicycle Stabilization : This thesis expands on the work of Randlov and Alstrom, using reinforcement learning for bicycle self-stabilization with robotic steering. The research included Deep Deterministic Policy Gradient (DDPG) algorithm training on virtual environments followed by simulations to assess its results. Furthermore, hardware testing was also conducted on Arizona State Universitys RISE lab Smart bicycle platform for testing its self-balancing performance.

• Achievement : Balancing time of 12-15 seconds was achieved on Smart Bicycle hardware system. Further improvements in regard to model training and hardware testing were also worked upon and need to be tested on the hardware system. MORE Student Researcher, Arizona State University, Tempe, AZ Jan 2019 - May 2019

• Research project proposal for alternate hydrogen gas storage solution to replace the compressed H2 storage in automo- biles for charging batteries faster than electric recharging. This research was to support RONN Motor Company.

• Achievement : 11 solutions were researched, out of which doped structure of Carbon Nanotube outperformed which was able to double the hydrogen carrying capacity of FCV (around 102%). PROJECTS

End to End Learning for Self Driven Cars

• Training a Convoluted Neural Network (CNN) to output steering commands for autonomous vehicle driving using training data collected from 3 camera setup on Udacity Unity simulator. Accuracy of 92% for testing data was achieved. Robotics Toolbox using App Designer in Matlab/Simulink

• Creating a user friendly toolbox for modeling and control of any robotic arm using Denavit Hatenberg (DH) conven- tions.The toolbox provides many functions that are useful for studying and simulating classical arm-type robotics. The module was successfully tested on low conguration robots and 6 DOF KUKA robot. Stationary balancing of bike using Control Moment Gyroscope (CMG)

• System Control knowledge applied on a stationary bike balancing project using Control Moment Gyroscope (CMG), designing controller using sliding mode control, dynamic modelling resulting in balancing of bike with reduced overshoot. EDUCATION

Master of Science, Mechanical Engineering (Robotics and Autonomous Systems) Aug 2018 - Dec 2020 Arizona State University, Ira A Fulton School of Engineering GPA: 3.67 Bachelor of Engineering, Mechanical Engineering Jul 2014 - May 2018 University of Mumbai, India GPA: 4.0

SKILLS

Robotics and Artificial Intelligence: C, C++, Java, python, OpenCV, Tensorflow, Tensorboard, Docker, doxygen, Git, Github, Supervisely, Pytorch, mongoDB, CosmosDB, Linux, Nvidia Jetson, Embedded Systems, ROS, Unity, Pan- das3D, Controllers, Sensor fusion, CUDA, Matlab, Simulink, MS Azure, Sensor technology, Rviz, Bash, AWS EC2 &S3 Concepts: Object Oriented Programming, Deep Learning, Robotics, Controls, Machine Design, Machine Learning, Data Structures, Algorithms, Agile/Scrum, Computer Vision Mechanical: Solidworks, CATIA, FMEA, Finite Method Analysis, AutoCAD, Pro-Engineer MOOCs / HACKATHONS

• Devils Invent + Banner Health Hackathon : Proposed a method for communication between Covid-19 patients and family to update on health status.

• Build a Self-Driving Car: Deep Learning and Computer Vision, Deep Learning Specialization: By Andrew Ng, Complete Python Course: Go Beginner to Advanced (7000 lines)

Contact this candidate