Prit Thakkar
I'm a computer Enthusiast, graduated from The University of Texas at Dallas with Master's degree in Computer Science with specializing in Data Science and was awarded Academic Excellence Award for Class of 2020. Recently, I did a year long Internship at Pacific Northwest National Laboratory as a Software Engineer Intern. During the Internship, I worked on multiple projects ranging from Web Development Project to Machine Learning Research Project working alongside Computational Scientists on COVID-19 data. Before that, I was interned at Indian Institute of Technology, Bombay working under Prof. Pushpak Bhattacharyya in the field of Natural Language Processing.
Work Experience
Pacific Northwest National Laboratory
Software Engineer Intern May'19-May'20
- Developed tool for Project managers and higher authorities to propose new projects in an organization and get approvals from the PNNL Directorate. Implemented MVC architecture with back-end in ASP.Net and front-end in knockout.js.
- Contributed to open source project under the supervision of computer scientist to enhance the features of a tool which uniquely identifies building footprints on Earth.
- Developed snakemake pipeline to preprocess molecular dataset, generate the refined dataset, and apply Machine Learning models on it. Used Docker to containerize the pipeline in order to test and analyze the result by domain scientists.
- Worked with Computational Scientists on COVID-19 Dataset to generate additional properties of molecules using the G-SchNet (Generative SchNet) model and MolDQN - a Deep Reinforcement Learning model.
- Developed 20+ Business applications over 2 months period used by approximately 4500 researchers at PNNL. These applications replaced old legacy systems decreasing overall maintenance cost by 45%
- Redesigned the database to follow modern database principles and automated complex workflows for these applications using MS Flow, Azure Service Bus and Azure Function reducing manual workflow by 70%
- Participated in daily meetings, sprint plannings and presented the project to CIO and other Senior Leadership Team members
UTD Computer Science Outreach
Student Instructor Nov'18-Present
- Taught programming language like JavaScript to school kids of 3rd to 5th grade in the after-school clubs.
Indian Institute of Technology, Bombay
Research/ Software Developer Intern Dec'17-Apr'18
- Built a RESTful API for "A WordNet based E-learning tool for Language Learning" using Django Framework to resolve the issue of Database Synchronization.
- Developed Relational Visualizer using D3.js and ConceptNet which displays different ontological relations of Hindi term using feature learning technique of NLP called Word Embeddings.
Jeavio
Software Developer Intern Dec'17-Apr'18
- Developed a Web-based application using PHP for generating attendance report of the employee of the company by parsing data from finger-print scanner.
Education
The University of Texas at Dallas
Master of Science-Computer Science
Specialization: Data Science
Coursework: Machine Learning, Design and Analysis of Computer Algorithms, Statistical Methods for Data Science, Web Programming Languages, Database Design, Big Data Management and Analytics, Natural Language Processing.
GPA: 3.96/4.0 | Transcript | Academic Excellence Award Winner
Aug'2018-May'2020
Dharmsinh Desai University
Bachelors of Technology-Computer Science & Engineering
Coursework: Data Structures and Algorithms, Discrete Mathematics, Advanced Java Programming, Object Oriented Programming, Software Engineering, Service Oriented Architecture, Artificial Intelligence, Data Mining, Web Technologies, Computer Networks.
GPA: 9.14/10.0 | Transcript
Jul'2014-May'2018
Projects
NLP-Template Extraction
NLP Techniques, NLP Pipeline, Machine Learning, Information Extraction
- Designed a NLP pipeline to perform Information Extraction task using NLP techniques on a given unstructured Wikipedia textual articles.
- Extracted templates like Buying event, part of relation between two locations, and job title relation between Person and Organization. Used various NLP techniques like WordNet features, Semantic Role Labeling, Dependency and Constituency parsing, Co-reference Resolution and NER.
- The proposed solution ranked top 10 in entire class.
Quick Draw: Doodle Classifier
Python, keras, CNN, AngularJs
- Built an Image based Classifier model which classifies hand-drawn sketches from 26 categories over 1 million dataset and achieved a 93% accuracy.
- Developed a local website where users can draw a sketch and system tries to predict the sketch.
Multi-Dimensional Search
Java, HashMap, Balanced Trees
- Developed a command line application in Java which searches a product using different attributes and descriptions just like E-commerce website.
Accounting Application
Python, Django
- Developed and deployed a web application on Heroku to manage accounts of different clients for Durga Petroleum - Gas Franchise.
EduAuto
Android
- Developed various modules of an Android Application such as attendance management and class announcements (news-feed) for an academic coaching class.
P2P File Transfer
Node.js, Peer.js, WebRTC, Heroku
- Designed a real-time browser-to-browser communication system for transferring files from one device to multiple devices without uploading files to server.
Restaurant Automation System
ASP.Net, MySql
- Designed a project which aims to automate restaurant management process with features like Table reservation, Menu Selection, Bill Payment and Membership.
