Portfolio Projects

Machine Learning, Deep Learning, & AI

Multimodal Zero-Shot Learning for Unseen Concepts - 2025

Designed a knowledge graph-based GCN model integrating CLIP text embeddings with semantic predicate-level relationships (including contrastive ones) to classify images without labeled training data; incorporated attention-based edge features and a custom prototype refinement loss and an MLP-based classifier head for final prediction.

PythonDeep LearningNeural Networks

Link

Reinforcement Learning for Abstractive Summarization - 2025

Designed a custom LSTM-based attention model for abstractive summarization on domain-specific dialogues (SAMSum dataset). Integrated reinforcement learning techniques (SCST, PPO) to improve fluency and context retention. Developed a modular training pipeline with a custom tokenizer and lightweight RL wrapper, no transformers required. Achieved 5.42% BLEU score improvement using PPO, with <3-hour training on modest hardware.

PythonDeep LearningNeural NetworksReinforcement LearningNatural Language Processing

Link

RePosture AI – 3rd Place, T9Hacks 2025

AI-powered posture correction tool using computer vision, ML, and real-time feedback algorithms.

PythonOpenCVMediapipeTensorFlowStreamlit

Link

LinkedIn Job Market Analysis

Developed a job market forecasting tool applying ML models for trend analysis and skill demand insights.

PythonTime SeriesClusteringRecommendation

Link

Full-Stack Development

Full-Stack E-Commerce Application

Developed a scalable, responsive web app with Google OAuth, state management, and payment integration.

React.jsNode.jsMongoDBExpress.js

Link

Java Based Expense Tracker

Designed and developed a full-stack expense tracking app with secure user authentication and automated reports.

JavaSpring BootMySQL

Link

Data Analysis & Visualization

Predictive Model of English Text

Developed a predictive language model using R Shiny and HC Corpora.

RShinyNLPn-grams

Link

NOAA Storm Database Analysis

Performed EDA on storm data using statistical modeling and data visualization.

Rggplot2dplyr

Link

Software Development

Console-Based ATM Interface

Built a Java-based ATM system with user authentication and transaction history tracking.

JavaOOP

Link

Published Research

CNN-based Cognitive Distraction Surveillance for Pedestrians using Phones and Headphones

Research paper on using convolutional neural networks to detect pedestrian distractions for improved safety.

Published in: 2023 International Conference on Control, Communication and Computing (ICCC)

DOI: 10.1109/ICCC57789.2023.10165546