macine-learning

Neural Dimensionality Tracker (NDT) is a production-ready Python library for high-frequency monitoring of neural network representational dimensionality during Neural Networks training

🎨 ArtGenre-Classifier: Machine learning library developed from scratch in C with Python interoperability (ctypes). Complete application for classifying artistic movements (Cubism, Renaissance, Surrealism) using Linear, PMC, and RBF models.

PHAGE — developed as part of Cybersecurity and AI coursework at Esprit School of Engineering. Real-time threat detection using machine learning and a multi-agent AI system.

An end-to-end serverless pipeline for Intelligent Document Processing (IDP) using Amazon Bedrock and Anthropic Claude 3 Sonnet. This project extracts structured data from scanned documents (e.g., PDFs, forms, invoices) using GenAI models, and stores results in a scalable cloud-native architecture.

Machine Learning based Monsoon Forecasting and Drought Risk Assessment system using Indian climate data, Random Forest classification, state-wise rainfall analytics, and interactive Streamlit visualization.

An AI-based accessibility solution that uses computer vision and voice assistance to detect objects, recognize currency notes and search for user-requested objects in real time, helping visually impaired individuals navigate their surroundings.

Bike rentals predictions by using live data from Api's with lightgbm regressor.

This project uses Logistic Regression on the Wisconsin Breast Cancer Dataset (30 features, 569 observations) to predict tumor malignancy, integrated into a Streamlit web app for real-time diagnosis.

An AI-powered tool that analyzes workflows, identifies inefficiencies, and suggests automation opportunities using Large Language Models.

Autonomous multi-agent Data Science pipeline — upload a CSV, get a trained model, EDA charts, and an executive report. Zero manual intervention. Built with CrewAI · FastAPI · React · XGBoost · WebSockets.

Integrated a real-time SLR system with a domain-specific LLM chatbot to enable natural, webcam-based communication for sign language users without the need for text or voice input.

Several implementations of machine learning models with accompanying reports. Covers both classical machine learning (regression, probabilistic learning) and deep learning (AlexNet, ResNet, etc.).

Extreme-aware AI framework for streamflow prediction and NWM bias correction using hydrometeorological data, temporal feature engineering, machine learning, and uncertainty-aware flood-event detection.

The Digital-Matters RAG Chatbot is an AI-powered assistant that allows users to search and explore the Utah Digital Newspapers collection using natural language questions.

A Decentralized Civil Registry Architecture leveraging Blockchain, Federated Learning, and Differential Privacy to secure National Identity Systems.

desktop application utilizes voice fingerprint and spectrogram technology to identify individuals based on their unique vocal characteristics.

A growing collection of my Data Science journey — notes, hands-on practicals, experiments, and projects across Python, SQL, statistics, machine learning, and beyond.

A simple project that aims to translate German text into English using a multi-head self-attention transformer model made with pytorch.

AI-powered Data Science Copilot for intelligent dataset analysis.

A supervised machine learning project that predicts student marks based on the number of hours studied using Linear Regression.