Hi, I'm Vivek Gupta.
A
Self-driven, quick starter, passionate programmer with a curious mind who enjoys solving complex and challenging real-world problems.
About
I am a recent computer science graduate from Purdue University. My interests span a broad range of fields, including Software Development, Machine Learning, Robotics and Computer Vision. I enjoy problem-solving and coding. Always strive to bring 100% to the work I do.
Prior to my graduate studies, I worked at Adobe for three years as Software Engineer II, where I learned valuable technical and professional skills. I am passionate about developing complex applications that solve real-world problems impacting millions of users.
- Languages: Python, Java, JavaScript, C, C++, Objective-C, HTML/CSS, Bash
- Databases: MySQL, PostgreSQL
- Libraries: NumPy, Pandas, OpenCV, scikit-learn
- Frameworks: Django, React, Node.js, Keras, TensorFlow, PyTorch, Bootstrap
- Tools & Technologies: Git, Docker, Kafka, AWS, GCP, JIRA
I’m currently looking for full-time Software Engineering or Machine Learning opportunities! If you know of any positions available, if you have any questions, or if you just want to say hi, please feel free to email me at gupta96v@gmail.com.
Experience
- Built framework for Adobe's unified extensibility platform using C++, Objective-C, Python and .NET, serving in-house teams and external developers to write 300 native plugins, leading to a 22.6% increase in revenue year-over-year ($2.88B)
- Collaborated cross-functionally to revamp code for 10+ core plugins integrating with scalable backend platform powered by JavaScript engine, improving applications' launch time by 50% and enhancing user experience for 15M customers
- Led team to build 20+ reusable React native design components, resulting in 3x reduction in features' development velocity
- Orchestrated the whole lifecycle management of new sharing feature inside Illustrator within a hard 6-month timeframe
- Received special contribution award for prototyping a Photoshop dialog module, resolving critical deployment bugs
- Tools: C++, Python, JavaScript, React, HTML/CSS, Node.js
- Collaborated to design a sequence model using LSTMs and NLP techniques to predict style breaches in documents
- Achieved 86% prediction accuracy surpassing the SOTA performance by 4 times, leading to the issuance of patent in US Tools: Python, NLP, Keras, TensorFlow
- Studied the responses of atmospheric gas sensors in various environment conditions using pattern recognition techniques.
- Prototyped a module to analyze the time-series data and predict gas concentration using an ensemble of supervised model.
- Tools: Python, MATLAB, Sequence Modeling, scikit-learn
- Created an image processing model in MATLAB that utilises NASA MODIS data to compute different crop monitoring indices like NDVI, WDVI, EVI, and SAVI for crop identification and detecting water body cover.
- Developed a satellite-based online information system for crop monitoring at district level.
Projects
Detecting Political Bias in News Media using contextualized embeddings
Predicting traffic flow and speed using Graph Multi-Attention Networks
Efficient Exploration using Bayesian Deep-Q Networks
Continuous Deep-Q Learning with Invertible Neural Networks
- Tools: Python, PyTorch, Generative Models, Reinforcement Learning
- Implemented off-policy deep-Q learning algorithm for continuous action space environments using the idea of invertible neural networks.
- Achieved 138% better average returns than Twin-delayed DDPG on OpenAI gym classical control tasks
A Convolutional Neural Network based Sign Language Character Recognition
- Performed Skin Segmentation using YCbCr model and morphological operations to segment hand from images with 90% accuracy
- Trained a convolutional neural network classifier using STL-10 database to recognize hand gestures
Group Recommendation using Inductive-Matrix Completion
- Built a novel recommendation system using tripartite sub-graph extraction and Relational Graph Convolutional Network (RGCN) to predict the preference ratings for a group-item pair.
- Obtained an RMSE score of 11.43 trained on 150,000 group-item and 100,000 user-item interactions for CAMRa2011 dataset.
Dynamic Adaptation of Software-defined Networks for IoT Systems
Skills
Languages and Databases
Python
C++
Java
HTML5
MySQL
Shell Scripting
Libraries
NumPy
Pandas
OpenCV
scikit-learn
matplotlib
Frameworks
Django
Bootstrap
Keras
TensorFlow
PyTorch
Other
Git
AWS
Education
Purdue University, West Lafayette
Indiana, USA
Degree: Master of Science in Computer Science
CGPA: 4.0/4.0
- Algorithm Design: Analysis and Implementation
- Information Security
- Data Communication and Computer Networks
- Compiling and Programming Systems
- Data Mining
- Reinforcement Learning
- Natural Language Processing
Relevant Courseworks:
Indian Institute of Technology, Roorkee
Roorkee, India
Degree: Bachelor of Technology in Electronics and Communication
CGPA: 8.73/10
- Fundamentals of Object-Oriented Programming
- Data Structures and Algorithms
- Database Management Systems
- Operating Systems
- Artificial Intelligence
- Computer Vision
Relevant Courseworks:
Teaching
An investment in knowledge always pays the best interest - Benjamin Franklin
I believe in the fact that knowledge increases by sharing but not by saving. It helps me to understand the concepts on a higher level, and motivates me and the students to appreciate the importance and impact of the subject. I have been a graduate teaching assistant to the following courses:
- Fall 2022, Spring 2023: Statistical Machine Learning (CS578)
- Spring 2022: Compilers: Principles and Practice (CS352)
- Fall 2021: Data Structures and Algorithms (CS251)