Chat with your PDFs, offline and privately. Unlock the knowledge hidden within your documents through a simple conversational interface. Built with multiple versions including local and React-based applications.

An end-to-end platform for email analysis and detection of smishing and vishing attacks, utilizing Flask, Next.js, and Java. Includes an Android application, interactive web app, and multiple machine learning models.

A web application that analyzes email content to detect spam. Uses multiple machine learning models (Random Forest, Logistic Regression, SVM, Neural Network, Decision Tree) to provide accurate predictions.

Trained a YOLO model on images to detect fire and developed a script to process videos, returning fire-detection outputs.

Built a deep learning model to predict stock prices using TensorFlow, Keras, Pandas, and LSTM neural networks. Explored time series data challenges and forecasting difficulties.