Hey, I'm
CS graduate who actually finishes things. I build across the whole stack — FastAPI and Flask backends, TensorFlow and PyTorch models, React and Next.js frontends, Docker containers, cloud deployments. Throw in some Three.js when things need to look good, and Tableau or Power BI when the data needs to speak for itself. Not chasing titles — just looking for work that's worth showing up for.
I built my first thing and got hooked. That's genuinely the whole origin story — no dramatic moment, no grand plan. Just the feeling of making something work that didn't exist before, and wanting more of that.
Since then I've been chasing that feeling across ML, web dev, data — whatever the problem needs. I care a lot about how things look and feel, not just whether they work. Probably comes from spending too much time gaming and noticing when interfaces are just right.
Right now I'm looking for the kind of work where I can keep getting better at the craft — a team that takes what they build seriously, and doesn't mind someone who asks a lot of questions.
0
Internships
0+
Certifications
0+
Projects Built
0+
Years Building
Full Stack Development Training
Research Intern — Computer Vision
Online Course — AI/ML for Geodata Analysis
Biostatistics Mentorship Program
hover a skill to highlight its axis
Biostatistics program centered on applying Machine Learning to complex biological and health data. It leveraged predictive modeling for insightful analysis and data-driven decision-making in life sciences.
Focused on using Artificial Intelligence and Machine Learning for analyzing geographical data. This involved applying advanced techniques for processing and interpreting geospatial information.
The internship applied deep learning, specifically UNet architecture, for image segmentation and polyp detection within medical colonoscopy images. It focused on advanced image processing techniques to aid in diagnostics.
Achieved certification in Oracle Cloud Infrastructure (OCI), covering foundational concepts of cloud computing services. This demonstrates a core understanding of cloud architecture, compute, and storage.
Completed a course focused on extending IBM DOORS Next functionality using JavaScript. This involved learning practical scripting for customizing and enhancing requirements management tools.
U-Net model achieving 95.20% precision on colonoscopy polyp detection — built during MNNIT Allahabad internship. Segmentation masks overlaid on colonoscopy images for clinical-grade accuracy.
Graph Convolutional Network for survival prediction achieving 85% concordance index — models entity relationships as graph edges for richer prognostic insight than classical ML.
CNN trained on the MNIST dataset for handwritten digit classification — with TensorBoard training visualisation and full model export for downstream deployment.
Binary classification predicting whether income exceeds $50K — Random Forest vs Gradient Boosting comparison on the UCI Adult dataset with full preprocessing pipeline.
Sentiment analysis on Amazon food reviews using NLTK VADER — 88% accuracy. Classifies reviews as positive, neutral, or negative and validates results against star ratings.
Interactive Tableau dashboard analysing Washington State EV adoption — breakdowns by manufacturer, model year, county, and CAFV eligibility.
DA Portfolio Site →Sales performance dashboard for a bicycle retail dataset — revenue, profit, product category mix, and regional performance with interactive Tableau filters.
DA Portfolio Site →Power BI dashboard on the Classic Models database — product-line revenue, order status tracking, customer segmentation, and DAX measures for YoY growth.
DA Portfolio Site →
Power BI analysis of the Netflix content library — genre distribution, release year trends, Movies vs TV Shows split, and top contributing countries.
DA Portfolio Site →
Normalised relational database for student academic records — schema design, stored procedures, and complex SQL queries for enrolment and grade management.
This portfolio — built with Next.js and TailwindCSS. Responsive design, framer-motion animations, modular component architecture, and a Resend email API.
Live Site →Full-stack web deployment of the Polyp Segmentation U-Net model — Next.js frontend for image upload and result display, FastAPI backend serving the trained TensorFlow model.
React.js web app where users draw a digit on a canvas and get a live CNN prediction — FastAPI backend serves the trained MNIST model for real-time inference.
A dedicated frontend website housing all data analytics projects — project cards with live Tableau and Power BI links, built in vanilla HTML/CSS/JS.
View Repo →Pixel-accurate Netflix UI clone with dynamic movie catalog and JavaScript-powered case-insensitive search — built in vanilla HTML/CSS/JS.
Android Tic-Tac-Toe in Java — GridLayout board, full game-state logic, player-turn tracking, and win detection compiled into a functional APK.
I'm currently looking for new opportunities, my inbox is always open. Whether you have a question or just want to say hi, I'll try my best to get back to you!
