📊 Analyze and visualize superstore sales data using Python, SQL, and Power BI to uncover insights that drive strategic business decisions.

✈️ | Predicting customer booking completion using Random Forest.

Big Data pipeline — PySpark, MLlib K-Means, Folium maps. Air quality analysis across 600 sensors in 68 countries.

News Article Semantic Classification | NLP pipeline for classifying news articles into 7 semantic categories

Semantic segmentation of onboard images in rural environments using an autoencoder architecture. Trained on the Yamaha-CMU Off-Road Dataset , with K-Fold validation and evaluation based on loss and mIoU.

The world's most accurate LLM. It achieves 0% hallucination rate by generating 0% of responses . the ultimate lightweight, bias free, zero-parameter machine learning architecture.

A precision-focused binary classifier that predicts diabetes onset risk from clinical metabolic markers, using predicted probability as a continuous progression proxy. Features fairness auditing across demographic groups, patient-level SHAP explainability, statistical drift monitoring, and a Streamlit clinician dashboard.

Data analysis, visualization, and machine learning practice using Python, Pandas, Seaborn, and Scikit-learn.

The **Advanced Financial Prediction Dashboard** is a comprehensive, state-of-the-art financial analytics platform designed to empower investors, traders, and data analysts with actionable insights and predictive intelligence.

End-to-end ML solution with Flutter frontend and FastAPI backend for intelligent car price estimation.

A sentiment classification application built with Streamlit, scikit-learn, and NLP preprocessing for movie reviews.

ML-Guided Compiler Flag Recommender (AutoFlag Finder)

FSM (Finite State Machine) + Task Manager Library for STM32

Localizing Memorized Sequences in Language Models

Reproducible pipelines comparing bulk RNA-seq, single-cell (pseudobulk), and scFM embeddings (scGPT, scFoundation) for per-drug response prediction in cancer cell lines.

AI-based cybersecurity system that analyzes network data to detect anomalies and potential threats in real time. Uses ML models for identifying suspicious patterns and ensuring basic system security as per the problem statement.

LegalGPT Pro is an AI-powered legal assistant designed for Indian legal professionals, law firms, and legal researchers. Built with Retrieval-Augmented Generation (RAG) technology, it provides precise, context-aware legal analysis and document-based question answering.

Real-time sentiment analysis tool for YouTube Live Chats, built with a Python FastAPI backend and a custom-trained machine learning model, and a TypeScript-based Chrome extension for the front end.

Spam vs Ham Detection using NLTK – A text classification project using Python, NLTK, and scikit-learn to distinguish spam from legitimate messages. Features balanced dataset handling, insightful visualizations, and performance evaluation with F1-score and confusion matrix.

This repository contains the code for traning, evaluating and deploying a model for dermatoscopic images of skin lesions.