About
Hi, I am Jannat — a passionate software engineer currently pursuing my Master of Science in Computer Scienceat Virginia Commonwealth University. I specialize in backend development, algorithms, and machine learning. I have worked with modern tools like React, React Native, Node.js,Python, and NLP frameworks to build scalable, efficient, and user-centric applications.
Whether I am optimizing data pipelines or crafting intuitive UI, I love solving real-world problems through tech. When I am not coding, you will find me exploring new frameworks or chasing down bugs like mastering the perfect serve in tennis — except bugs don't usually cooperate as nicely as tennis balls!
Experience

Research Assistant
Natural Language Processing Lab, Virginia Commonwealth University
- Preprocessed 10,000+ biomedical abstracts using Python for clean, tokenized inputs, boosting model efficiency by 30%.
- Contributed to a relation extraction model using PyTorch, improving F1-score by 12% through architecture refinement and training.
- Used TorchMetrics to automate evaluation across 18 relation classes, reducing manual validation time by 50%.

Graduate Student Software Engineer
Virginia Commonwealth University
- Developed a student-alumni mentoring interface using Google Sheets and Apps Script for Career Services.
- Implemented features such as mentor profile viewing and automated email delivery, enhancing user experience.
- The platform is now being used by more than 100 students, significantly improving engagement and accessibility.

Teaching Assistant
Virginia Commonwealth University
- Mentored 20+ students weekly in C programming labs, guiding them in debugging, data structures, and algorithmic thinking.
- Supported 20+ first-year students in Engineering 101 with fundamentals in Arduino, circuit theory, and theremin projects.

Software Engineering Intern
Raceme Tenders LLP
- Upgraded legacy PHP codebase for 5+ modules and successfully deployed to AWS via SSH, reducing server downtime by 20%.

Software Engineer Intern
Raceme Tenders LLP
- Built an automated Python crawler with Selenium to extract 1000+ tenders from the Karnataka e-Procurement portal in collaboration with one teammate.
Recent Projects


ragResearch
AI-powered research blog platform with real research paper integration, vector search, semantic understanding


GutBrainIE
Designed a robust relation extraction pipeline focused on gut-brain biome axis literature.


Tennis Analysis
Analyzes tennis videos for player & ball tracking and court line detection using modern computer vision.


Carpooling App
Seamlessly connect with colleagues and friends who needs a ride and those who can provide ride to the campus.
Tech Stack
Next.js
React.js
Node.js
Express.js
JavaScript
Tailwind CSS
HTML5
CSS3
PostgreSQL
MongoDB
Firebase
MySQL
Python
TensorFlow
PyTorch
Scikit-Learn
Hugging Face
Pandas
GitHub
Postman


