Dev Shah

About

I am a skilled software engineer with expertise in full-stack development, machine learning, robotics, and autonomous systems. With more than 2 years of experience in building software applications, my work includes designing dynamic websites, deep learning models and cloud-native technologies, improving precision and efficiency across various projects. Adept at applying first principles thinking to solve complex problems, leveraging strong data structures and algorithms expertise. Passionate about solving complex problems with innovative, data-driven solutions.

Degree: Master of Science
Major: Computer Science

Email: shahdev99@gmail.com
LinkedIn: linkedin.com/in/devshah7

Download my CV

Skills

All
Languages
Frameworks
Tools

JavaScript 100%

Python 90%

Java 75%

MATLAB 60%

C++ 55%

Julia 45%

React JS 100%

Scikit-Learn 90%

Pandas 90%

YOLO 70%

OpenCV 60%

ROS/ROS2 45%

Git 80%

Gazebo 65%

Simulink 55%

Choregraphe 45%

* confidence level

Resume

Education

Master of Science in Computer Science

August 2022 - May 2024

The George Washington University, Washington, DC, USA

Machine Learning, Computer Vision, Robotic Systems, Cloud Computing, Tele-Medical Robotics, Secure Autonomous Systems

Bachelor of Engineering in Computer Engineering

August 2017 - May 2021

Atharva College of Engineering, Mumbai, MH, India

Software Development, Artificial Intelligence, Natural Language Processing, Human-Machine Interaction

Professional Experience

Research Assistant

February 2023 - December 2023

ART-Med Lab, The George Washington University, Washington, DC, USA

Enhanced precision in microrobot swarm simulations by 25% by engineering a haptic feedback system with Chai3D and OpenCV.
Increased autonomous system navigation capabilities by 30%, achieving fewer navigation errors, by designing a probabilistic obstacle avoidance algorithm.

Software Developer

January 2021 - July 2021

Depronto Infotech, Mumbai, MH, India

Developed twelve banking applications, leading to high user adoption and performance, by leveraging ReastJS to a user-centric design
Boosted software testing efficiency by 40% by developing an automated testing platform using Cypress and NextJS
Provided mentorship to trainees, focusing on problem-solving, documentation, and collaboration initiatives

Portfolio

All
Apps
A.I.
Robotics

JD Archi-Terior Designs site

This freelance work showcases a portfolio of architectural and interior design works. It features detailed sections on various architecture and interior projects like educational institutions, residences, cafes, and thematic museums. The website illustrates innovative design solutions tailored to client needs for each project, emphasizing functionality and aesthetic appeal.

HTML
CSS
JavaScript

V2V Misbehavior Detection

This project focuses on developing a misbehavior detection system for vehicular ad hoc networks (VANETs) using machine learning. It involves data collection via simulation, preprocessing, and testing various ML algorithms like KNN, RNN, CNN, SVM, and one-class SVM to improve traffic safety and efficiency.

PYTHON

Quadrotor Control System

This project involves designing and implementing a quadrotor control system using MATLAB, developing a mathematical model based on aerodynamics and Newtonian mechanics, and employing PID controllers to manage flight modes, tested via simulations.

MATLAB

Rover Control & Attack

Designed and implemented attacks on a rover's control system using ROS, Python and C++ to assess vulnerabilities, manipulate PID controls, and disrupt mission execution, enhancing skills in robotic control systems, mission planning, and security assessment.

ROS
C++
Python

Strawberry Ripeness

This project utilizes the YOLOv8 deep learning model to classify the ripeness and developmental stages of strawberries. By implementing data augmentation techniques and training on a diverse dataset, it achieved a mean average precision (mAP) of 82.1%, enhancing accuracy in agricultural practices and crop management.

YOLO

Genetic Algorithm

This project simulates virtual creatures that evolve and learn to efficiently move and gather food using a genetic algorithm to adjust the weights of a neural network, reflecting the "survival of the fittest" principle. It demonstrates evolutionary strategies and neural computation for autonomous decision-making in dynamic environments.

JULIA

Basic Machine Simulator

This project developed a basic machine simulator to enhance understanding of computer systems, focusing on CPU macro- and microstructures, instruction set architecture, and memory handling. The simulator, designed in a CISC architecture style, provides interactive tools for visualizing and manipulating computer operations.

JAVA

Contact

Location:

Washington, D.C.

Email:

shahdev99@gmail.com

LinkedIn:

linkedin.com/in/devshah7

Github:

github.com/devshah7