Hello There! I am

Adarsh Thoke

I'm a

About

About Me

I am a recent graduate in Computer Science from the University of Colorado, Denver, with over two years of experience specializing in full-stack software engineering and machine learning systems. I am currently seeking new opportunities to apply my skills in impactful projects. During my graduate studies, I had the privilege of working under the guidance of Prof. Ashish Biwas in the field of machine learning. Additionally, I collaborated with Prof. Liang He on ARADISS (Adaptive Real-time Anomaly Detection and Identification for Space Systems), a project funded by NASA. My previous research includes Wind Turbine Anomaly Detection, where I focused on problem formulation and data-driven approaches, identifying the most correlated variables and building predictive models using SCADA data.

  • Name: Adarsh Thoke
  • Date of birth: December 06, 1998
  • Address: New York, NY, 07306 USA
  • Email: adarshthoke.us@gmail.com
  • Phone: +1(949)-615-9700

0 Project complete

Download CV

Education

2022-2024

Master of Science in Computer Science

University of Colorado Denver GPA: 3.67

Relevant Coursework: Machine Learning, Computer Vision, Big Data Science, Artificial Intelligence, Deep Learning, Advance Computer Architecture, Operating Systems and Algorithms.

2022-2024

Bachalor of Science in Information Technology

University of Mumbai GPA: 8.0/10

Experience

2023-2024

Machine Learning Intern

Global Technology Connection Inc. and University of Colorado, Denver.

  • Led ARADISS (Adaptive Real-time Anomaly Detection and Identification for Space Systems), a NASA funded project focused on classifying anomalies in battery-based vehicles.
  • Collaborating with business stakeholders Global Technology Connection, Inc. to analyze large amounts of data to discover complex patterns and employ a data-driven approach.
  • Designed an algorithm for differentiating Dependent (Cascaded) and Independent (Non-Cascaded) anomalies using insights and ML XGBoost model extracted new features using statistics and advanced analytics. Currently, in the testing phase, the model demonstrates an accuracy ranging from 92% to 96%.
  • Applying data visualization and manipulation techniques for intuitive communication and presentations.
  • Previouly work on Wind Turbine Anomaly Detection. Did Abstraction of variables on a graph model(To figure out which variables to use) as well as develop models to simulate it.

2020-2022

Software Engineer

Reliance Jio Platforms Ltd, Navi Mumbai, India.

  • Collaborated with the cross-functional Enterprise team at Jio, developed new features, and crafted APIs for the Customer Management Portal for sending promotional templates to more than 400M plus users. Provided development Support for Knowledge Management System.
  • Created streamlined access to fetch keys from WhatsApp(Meta) server after uploading Jio promotional Template from scratch and deployed it. Implemented Kafka which optimizes user query distribution and Redis to make the system scalable, overall improving the performance of the system. Additionally, prepared comprehensive Technical Documentation for the entire system.
  • Worked with the Customer Experience Team to improve the product analytics and experience of the live chat functionality of JioCare on WhatsApp, HelloJio and JioCare in MyJio with a user-centric approach.
  • Defined product-specific KPIs, tested 2000+ queries, and addressed critical blockers. Implemented business strategies, boosting customer acquisition by up to 8% in three months.

Skills

CSS

90%

Python

95%

HTML5

95%

Java

90%

React

75%

Kafka

70%

Redis

70%

JavaScript

80%

Docker & Kubernetes

80%

AWS & Azure

60%

TensorFlow

75%

ML/AI Technologies

60%

Data Structures & Algorithms

80%

Projects

June 2024

E-Commerce Website using Python and Django

GitHub Link

• An end-to-end e-commerce website with 3 core functionalities like user identification, product creation, and cart functionality.
• Implemented features like user session management, checkout process, wishlist, and price filtering, enhancing the platform’s user experience and serving over 1,000 users.

May 2024

Banking Transaction System

Pvt. GitHub Link

• Created a transaction system using Java and the Spring Framework, adhering to MVC architecture.
• Implemented CRUD operations to manage transaction data and integrated event logging capabilities to ensure 100% accurate tracking and auditing of all transactions.

January 2024

Student Performance Indicator

GitHub Link

• Trained 5 different models like Decision Tree, K-nearest neighbors, boosting, and bagging models, performed EDA, data transforma- tion, and designed a data pipeline deployed on Microsoft Azure.
• Achieved optimal model selection using R-squared (R2) metrics to capture variability in student performance data

January 2024

Smart Vehicle Assistance System.

GitHub Link

• Develop Web App for Traffic Signal recognition, employing a CNN model via Keras. Backend powered by Python, and Flask, enabling model interaction. Leveraged ReactJs for a responsive frontend.

August '23 - December '24

Hand Gesture Control System.

GitHub Link

• Used OpenCV for real-time video frame capture from a webcam, implementing image processing functions for preprocessing. Integrated Mediapipe for hand tracking, leveraging pre-trained models to extract 3D hand landmarks. Used Euclidean distance calculations between hand landmarks, enhancing finger point analysis.

August '22 - December '23

Dynamic Pricing of food products to reduce wastage

GitHub Link

• Conceptualized pricing reduction strategy using parameters like expiry dates to enhance sales and minimize discarded inventory.
• Generated a dataset with 1000+ different food products, did data mining, and data modeling using supervised learning (Linear Regression).

August '22 - December '23

Chain Reaction Game

GitHub Link

• Designed a working prototype of a chain reaction game using a minimax algorithm with alpha-beta pruning.
• Technologies used: Python and Pygame libraries.

Undergrad '20

Human Detector Robot- Drishti.

• A working prototype of a snake bot that was funded by the United Nations Development Programme and Navi Mumbai Municipal Corporation. Designed to find humans under any natural calamity. The technology used Arduino UNO, a C language to code and control the movement of the servo motor.

Undergrad '19

Object Recognition - Reckon.ai

• Implemented an AI-based application for object detection.
• Technologies used: Caltech 101 dataset, OpenCV, TensorFlow, and Android Studio.

Undergrad '19

Smart Toilet Feedback System

• Designed an IoT-based project for collecting cleanliness feedback from public toilets via a display system.
• Prototype included sensors to count entries and an automated system for the toilet.
• Technologies used: Arduino UNO, IR sensor, Relay Motor.

Awards

2017 - Present
  • Distinguished Graduate Master’s Capstone Award for research from the University of Colorado Denver. 2024
  • Stanford University & DeepLearning.AI: Machine Learning Specialization. 2024
  • Udemy: Python for Data Science and Machine Learning Bootcamp. 2023
  • Delivered a Code Refactoring seminar to 120+ students, selected by the Head of IT Department. 2021
  • Won a cash prize of 50,000 INR from Navi Mumbai Municipal Corporation and United Nations Development Program for developing a snake bot for natural calamities. 2019
  • Organized TEDxDYPatilUniversity event, served as "Events Head" for the Annual College Cultural Fest. 2017-2019