A showcase of my technical projects and applications
A specialized implementation for glucose time series forecasting using Large Language Models (Time-LLM framework) for Type 1 Diabetes management. This project leverages advanced LLMs (GPT2, LLAMA, Deepseek) to predict future glucose levels based on historical measurements, insulin administration, and meal information. Features multi-model support, comprehensive evaluation metrics (MAE, RMSE), and achieves 21.87% reduction in prediction error compared to state-of-the-art models with 96.19% clinical accuracy.

A comprehensive news research application powered by LangChain for advanced natural language processing. This tool seamlessly integrates the OpenAI API to generate concise summaries, perform in-depth analysis of news articles, and identify key entities. It also features a user-friendly interface for searching, filtering, and organizing news data, enabling users to stay informed and gain valuable insights from a vast amount of information.

A perfect mangament tool for managing the hotel and the services offered by it. Comes up with the various features such as room service, checkin, checkout, customer service, etc.

Postman clone is created with the help of React and Material UI. It contains GET and POST method. The three fields Params, Body and Header is included in this project.

This is the responsive webiste of civil service company named "SSk Civil Works and Services" created with the React.

This project is a chatbot application that utilizes Gemini models from Google Generative AI. It provides a user-friendly interface for interacting with the chatbot and supports both text and voice inputs. The chatbot can handle various configurations and offers enhanced features like voice recognition, text-to-speech responses, and sentiment analysis.

This project implements a simple Distributed Product Management System using Java. It follows a client-server architecture where: A NameServer keeps track of different product tables and their respective servers. Multiple Database Servers store product data dynamically. A Client interacts with the system to insert, delete, and query products.

A sophisticated medical chatbot leveraging LangGraph, LangChain, and Google's Gemini model to provide accurate medical information through a multi-agent system. Features Multi-agent architecture for intelligent query processing Medical knowledge base integration via Pinecone vector store Real-time web search capabilities Natural language understanding using Google's Gemini 1.5 Flash Interactive Streamlit web interface

This project is an AI-powered chatbot that interacts with the ThingSpeak API to retrieve temperature data based on user queries. The chatbot is designed to work with the Gemini 1.5 Flash API and provides a Streamlit dashboard for user interaction.