Projects

My Projects

I love to Build Cool Projects. Here, you'll find a curated collection of my creative endeavors and technical projects. Each piece represents a journey of innovation, problem-solving, and continuous learning. Feel free to explore this showcase of my passion and expertise in action.

PhotoClassifier – On-device photo organization with AI

PhotoClassifier is a native iOS/macOS app that uses Core ML and CLIP vision embeddings to classify and cluster photos into albums based on visual similarity, all processed securely on-device.

SwiftCore MLCLIPPhotos FrameworkiOSmacOS

Stock Transformer – Time series modeling for financial prediction

Developed a transformer-based model for stock return prediction using historical pricing and factor data, with a focus on minimizing ranking error via the Rank Information Coefficient.

PythonTransformersPandasKaggleFinance

Fantasy Daily Leaders

Fantasy Daily Leaders is a React-based web application designed to track and display top-performing fantasy sports players on a daily basis. Built with React, styled-components, and Axios, this project fetches and presents real-time or scheduled player data in a modern, user-friendly interface. Ideal for fantasy sports enthusiasts seeking quick insights and daily performance updates.

ReactFantasy SportsWeb AppStyled ComponentsAxios

RNA 3D Folding – Structure prediction with deep learning

Built a transformer-GNN pipeline for predicting RNA tertiary structure for the Stanford RNA 3D Folding challenge, integrating RibonanzaNet for accurate TM-score evaluations.

PythonTransformersGNNTM-scoreKaggle

MovieLens Recommender – Personalized movie suggestion engine

Created a React and Python-based movie recommender system that uses collaborative filtering to suggest films based on user ratings of 10 movies.

ReactPythonMovieLensCollaborative FilteringRecommender Systems

Drawing with LLMs – Text-to-SVG generation system

Participated in the Kaggle 'Drawing with LLMs' competition, creating a package that converts natural language prompts into valid SVG code using generative models.

PythonSVGNLPGenerative ModelsKaggle

Portfolio Site – Next.js developer portfolio from Figma design

Built a pixel-perfect developer portfolio website with Next.js 14, closely replicating a Figma design with custom color schemes and fonts.

Next.jsTailwind CSSFigmaReact