Hello, I'm Vandit
Currently specializing in Computer Science, with a passion for Machine Learning and Computer Vision
Introduction
Overview
Hello there! I'm a dedicated Data Scientist deeply engaged with Cloud and Big Data tech, particularly Microsoft Azure. My journey includes leading teams and enhancing data management and analytics for Fortune 500 clients in education and public sectors as a Digital Specialist Engineer at Infosys. Currently expanding my horizons with a Master of Science in Computer Science at Northeastern University, I've dived into Machine Learning, Data Analytics, and Computer Vision. My research has led to four publications, focusing on text classification, predictive analytics, and accident detection. I'm all about solving real-world problems with image processing and computer vision, driven by a blend of innovation and efficiency. Let's connect and explore how we can innovate and revolutionize the digital landscape together!
Data Scientist
Data Engineer
Software Developer
Web Developer
What I have done so far
Work Experience
Data Engineer
Infosys Pvt. Ltd- Managed an 8-member cross-functional team to build highly efficient Azure ADF ETL pipelines, resulting in a 50% improvement in data extraction speed from diverse external sources and streamlining the data ingestion process
- Transformed client's codebase from USQL to TSQL using Azure SQL, optimizing query execution time by 40%, and saving 150+ man-hours monthly in data processing. Infrastructure upgrades also implemented to boost efficiency and reliability
- Boosted data access speed by 30% from Azure Gen2 data lake storage for users and increased data processing by 40% using efficient PySpark and TSQL scripts for Change Data Capture (CDC), tailored for real-time streaming data
- Streamlined code deployment with Agile methodologies, Shell scripting, Git for version-control, and Azure DevOps CI/CD integration, resulting in a 20% testing time reduction and a 15% decrease in post-deployment issues
Data Analyst Intern
Spotle.ai- Performed Exploratory Data Analysis (EDA) and feature engineering on clickstream data using NumPy, Pandas, and Scikit-Learn, optimizing analysis with AWS Redshift for enhanced performance and data accessibility
- Increased quarterly revenue by 10% and user engagement by delivering KPI-focused Tableau dashboards, written reports, and marketing insights to stakeholders, accelerating data-driven decision making for student content optimization
- Streamlined user and student data processing by utilizing AWS Glue and S3 bucket, achieving a 23% reduction in ETL processing time, resulting in enhanced scalability, efficiency, and cloud cost savings
- Implemented task coordination using Agile project management tools like Slack and JIRA for organized project tracking
Artificial Intelligence Intern
Allsoft- Enhanced user engagement and response accuracy by 40% by developing and implementing a sentiment analysis REST API using Flask and a Bi-LSTM model, seamlessly integrated into the client's platform's chatbot
- Refined the client’s recommendation engine, achieving a 15% boost in user engagement, by utilizing item-based collaborative filtering to improve product recommendations for customers
Technologies
Technical Skills
My work
Projects
Following projects showcase my skills and experience through real-world examples of my work. Each project is briefly described with links to code repositories and demos. It reflects my ability to solve complex problems, work with different technologies, and manage projects effectively.
Image Editor
Developed a Java Image Processing application with Java Swing for the UI, applying OOD principles like MVC and Builder patterns to ensure maintainability. Version control facilitated team collaboration and code stability. The UI was designed for easy navigation, optimizing user experience, while JUnit ensured robust functionality. The Builder pattern was key to streamlining feature development, cutting implementation time by 70%.
#java
#swing
#junit
EventBuzz
Led 'EventBuzz', significantly enhancing event planning efficiency by 30% through the implementation of CRUD operations, and improving reliability by 25% using MySQL and Node.js. Integrated advanced logging mechanisms, achieving a remarkable 99.9% uptime. Employed React for creating a dynamic, user-friendly front end, contributing to a robust and adaptive event management platform. These enhancements streamlined coordination and adapted to user needs.
#mysql
#react
#node
Eazy_Park
Led team Coding_Capitals3 to 2nd at Student Hackday 2019, Delhi, innovating a parking solution with Masked R-CNN and COCO dataset. Achieved 0.92 confidence in detecting 80 types, enhancing urban parking spot identification and accessibility. The project demonstrated deep learning expertise and a commitment to addressing urban challenges, contributing to smarter city solutions and improving daily life with advancements in object recognition.
#python
#tensorflow
#flask
Research
Papers
Following projects showcase my research work and demonstrate my ability to explore innovative solutions and tackle complex problems. Each project is briefly described with links to research papers and related materials. It reflects my dedication to advancing knowledge in my field of interest.
Machine Learning enabled models for YouTube Ranking Mechanism and Views Prediction
Designed and implemented a predictive model using machine learning and AI to estimate YouTube video metrics, incorporating a ranking system based on trending data. Enhanced the model with a Random Forest algorithm, achieving a notable 78.4% accuracy, aiding creators in evaluating content authenticity and competitive strategy before publication
#analytics
#tensorflow
#youtube
Classifying text using machine learning models and determining conversation drift
In this research, Naive Bayes, Logistic Regression, and SVM algorithms were optimized for off-topic text detection, achieving an F1 score of 0.81. This significantly advances discussion moderation techniques, enhancing accuracy and efficiency. The work contributes to better engagement in digital conversations and suggests further applications in language and moderation technology.
#python
#nlp
#deepLearning
Accident Detection Using Mask R-CNN
The paper presents the Mask R-CNN model as a viable, efficient tool for accident detection, utilizing deep learning for accurate instance segmentation. The model uses the Intersection over Union concept to precisely detect and identify accidents on roads. By effectively identifying mishaps, it aids in averting traffic jams and expediting emergency response, which could save lives and crucial time. The Mask R-CNN model's accuracy and potential to integrate with response systems mark significant advancements in traffic safety and management.
#cnn
#keras
#objectDetection
My Achievements
Honors & Awards
Below are some of my professional and academic achievements and honors that I have received over the years.
Rookie of the Quarter
- Received the 'Rookie of the Quarter' award at Infosys for exceptional performance and significant contributions to project success, rapidly adapting and boosting team efficiency.
Insta Award
- Earned the 'Insta Award' at Infosys for mastering Azure tools like Data Factory, Data Lake, and Synapse Warehouse, excelling in project development, coordination, and timely delivery.
ZS Campus Beats Challenge
- Secured a spot in the Top 11-100 teams in the ZS Campus Beats Case Challenge 2020, earning an interview for Business Technology Analyst at ZS and advancing to Phase-3.
Best Student in Computer Science
- Honored as the best Computer Science student in 12th grade at NC Jindal Public School, recognizing my academic excellence and leadership in the field.
Academic Excellence Awards
- Received the Scholar's Badge and Certificate of Scholarship for outstanding academic achievements throughout 11th and 12th grade.
Published Research
- Published four impactful research papers in machine learning, including innovative works on text classification, YouTube ranking, COVID-19 detection, and accident identification using advanced deep learning techniques.
What others say
Testimonials
"
Vandit never stopped asking questions from the day he joined the team. I have worked with him for a year where he has shown tremendous ability to learn new skills. He was one of the core members in the team, who was involved in the successful implementation of a re-engineering program in support of the latest azure cloud offering. I highly recommend Vandit to any organisation.
Lead Data Engineer of The Open University, United Kingdom
"
I wholeheartedly endorse Vandit Gupta for his outstanding academic and research contributions. His innovative approach, leadership in collaborative projects, and significant conference papers highlight his deep understanding of technology and research. Vandit's dedication and technical acumen make him an exemplary candidate for advanced endeavors.
Assistant Professor of Maharaja Agrasen Institute of Technology, India
"
Vandit's passion, proficiency, and problem-solving skills make him a standout team member with a rare blend of technical prowess and interpersonal skills. At Open University, he's been pivotal in data solutions, displaying deep knowledge in data pipelines and ETL processes. His commitment to cooperation and continuous learning makes him an invaluable asset.
Data Engineering Lead of The Open University, United Kingdom
Get in touch