Machine Learning Engineer, Data Scientist and Roboticist

KUSHAGRA AGRAWAL

******************@*****.*** +1-240-***-**** linkedin.com/in/kagrawal97 github.com/kushagra7176 EDUCATION

UNIVERSITY OF MARYLAND College Park, MD

Master of Engineering in Robotics Expected May 2021 GPA – 3.87/4

Relevant Coursework: Machine Learning, Control of Robotics Systems, Computer Vision, Decision Making for Robots, Data Science, Software Development for Robotics

MANIPAL UNIVERSITY Manipal, India

Bachelor of Technology in Mechanical Engineering May 2019 SKILLS

Programming Language(s): Python, C++, R, MATLAB

Tools and technologies and OS: Linux, Windows, OpenCV, Scikit-learn, PyTorch, TensorFlow, Kera’s, ROS, GIT, MySQL, MS SQL, PostgreSQL, MongoDB, AWS Sage maker, MS Office, CMake, Valgrind, Travis, Gtest Design and CAD Tools: Auto CAD, CATIA, ANSYS, Simulink, V-rep TECHNICAL EXPERIENCE

Robert Bosch India Limited Bangalore, India

Mechanical Design Intern Feb-July 2019

• Designed and updated the throttle body of diesel and petrol engines using AutoCAD and CATIA which was used in the final catalogue.

• Validated throttle body of diesel engines in a team of 3 using Computational Fluid Dynamics (CFD) analysis. Bharat Heavy Electricals Limited Bhopal, India

Mechanical Intern June- July 2018

• Assisted the engineers in the assembly of the Francis turbine and performed operations such as facing, step turning on lathe.

• Performed minor machine maintenance and determined machine adjustments and material requirements. PROJECTS

Human Obstacle Detector Nov 2020

• Developed a C++ based software for real time object and tracking using monocular camera.

• Agile Iterative Process (AIP) guidelines and Test-Driven Development (TDD) process were used to develop the module.

• Ensured the quality of the software by writing unit tests using continuous integration. Dog Breed Classifier July 2020

• Built a CNN model in TensorFlow and used it to predict the breed of a dog. Refined the model using transfer learning.

• Used Resnet101 architecture for transfer learning which increased the accuracy from 72% to 91%. Path Planning for a Turtle-bot May 2020

• Implemented Dijkstra, A* and RRT* algorithm for finding path between start and goal point while avoiding obstacles.

• A custom shape map was build using python and the final output was shown in a GUI drawn using OpenCV in Python. Visual Odometry April 2020

• Calculated Fundamental and Essential matrix using 8-point algorithm and improved the results by integrating RANSAC and Zhang’s estimation method.

• Calculated camera pose using linear triangulation while checking for chirality condition. Advanced Lane Line Detection using Python and OpenCV March 2020

• Used concepts of color segmentation, homographic transformations and sliding window algorithm to develop a code that can detect lane lines from the noisy image frames in 98% of the frames. AR Tag detection Feb 2020

• Detected the corner and orientation of the tag using OpenCV.

• Calculated the homography and projection matrix and used them to superimpose an image and cube on the tag. Controls of Two Pendulum Cart Setup Nov-Dec 2019

• Developed an LQR and LQG controller for the two-pendulum cart setup and simulated the results using MATLAB. Maze Solving Robot Nov-Dec 2019

• Developed a C++ program to drive a robot inside the maze from the start position to goal position.

• Breadth-First Search algorithm was used to compute the path and navigate the robot inside the maze. PROFESSIONAL CERTIFICATION

Udacity Machine Learning Engineer Nanodegree July 2020

• Deployed Sentiment Analysis model, Payment Fraud detection, Plagiarism Detector model using AWS Sagemaker.

• Developed and deployed a predictive analysis model to determine whether a person is probable customer or not based on the demographic data given by Arvato Financial Services.

• Used XGBoost, AdaBoost, Linear and Logistic Regression, K Means and other Machine Learning algorithms in the process.

Contact this candidate