VIVEK GUPTA

• 765-***-**** • adylkt@r.postjobfree.com • guptav96.github.io •guptav96

Education

Purdue University, West Lafayette Indiana, USA

Master of Science in Computer Science Aug 2021 – Aug 2023

GPA: 4.0/4.0

Coursework: Data Structures and Algorithms, Compilers, Statistical Machine Learning, Network Security, Computer Networks, Database Management, Object-Oriented Programming, Natural Language Processing

Indian Institute of Technology, Roorkee Roorkee, India

Bachelor of Technology in Electronics and Communication (minor: Computer Science). Jul 2014 – Jun 2018

Skills

Programming: C, C++, C#, Python, Objective-C, Java, Scala, JavaScript, HTML5 and CSS3

Machine Learning: NumPy, Pandas, SciPy, scikit-learn, Matplotlib, Jupyter, PyTorch, OpenCV, NLTK, TensorFlow

Libraries / Frameworks: React.js, Node.js, .NET, Spring Boot, Django

Software / Technologies: XCode, Git, Docker, Kubernetes, Grafana, ROS, Apache Kafka, Jira, Jenkins, A/B Testing

Database: MySQL, PostgreSQL, NoSQL (Neo4j), Elasticsearch

Cloud: AWS, Azure, GCP

Experience

Purdue University, West Lafayette Aug 2021 – Present

Graduate Research Assistant, Graduate Teaching Assistant

Developed containerized microservices architecture with Docker to build digital forensic tool for a confidential US agency, improving resource utilization by 85% and facilitating seamless integration of React.js frontend with RESTful APIs.

Architected and executed an advanced value-based RL policy using invertible neural networks; driving an exceptional 138% increase in returns compared to DDPG on OpenAI gym tasks.

Instructed 80+ graduate students as a teaching assistant for statistical machine learning and compilers course.

Adobe Systems Jul 2018 – Aug 2021

Software Engineer II

Built scalable backend for Adobe’s extensibility platform framework, serving 8 in-house teams and over 300 contributors.

Developed integration APIs and successfully integrated Adobe products with UXP framework, enabling cross-platform compatibility and improving launch time by over 50% resulting in an improved user experience for 15M customers.

Led team of five in successful development of a new sharing feature for Adobe Illustrator within a tight 6-month timeline.

Collaborated cross-functionally to design and implement user interfaces for 10+ components, including Cloud Documents and Neural Filters for Adobe flagship products.

Big Data Experience Lab, Adobe Inc. May 2017 – Aug 2017

Research Intern

Proposed a deep learning model utilizing LSTMs and NLP techniques to predict and mitigate style breaches in documents.

Achieved an impressive 86% prediction accuracy rate, leading to the issuance of a patent in the United States.

Publications

Vivek Gupta, Praphpreet Dhir, Jeegn Dani, and Ahmed H. Qureshi. Maner: Multi-agent neural rearrangement planning of objects in cluttered environments.

Vivek Gupta, Pranav Ravindra Maneriker, Anandhavelu Natarajan, et al. Predicting style breaches within textual content, May 12, 2020. US Patent 10,650,094.

Vivek Gupta, Naresh Kumar, Aditi Sharma, and Ajith Abraham. Sensor routing protocol with optimized delay and overheads in mobile-based wsn. Journal of Information Assurance & Security, 16(4), 2021.

Projects

Multi-agent Rearrangement Planning: Pioneered creation and application of task and motion planning framework using vision transformers for multi-agent object rearrangement. Demonstrated exceptional results, surpassing conventional methods with success rate improvement of 20%. (Work in review at IEEE Robotics and Automation Letters).

Group Recommender: Introduced a novel recommendation system for estimating group preference ratings, utilizing a tripartite sub-graph extraction and Relational Graph Convolutional Network (RGCN) model; achieved an RMSE score of 11.43 for movie rating dataset.

Bayesian Deep-Q Networks: Implemented Bayesian Deep Q-Networks in PyTorch and conducted experiments on 8+ atari games; outperformed Double DQN in training time and improved returns (by median of 300%).

Contact this candidate