This File is a summary of AthulKrishna's Projects. Project 1: name:System Metrics Dashboard A sleek, interactive web dashboard built with **React** and **Recharts** for monitoring real-time system performance metrics like CPU usage, memory stats, and custom metric tracking. Designed to be modular, responsive, and easily integrable with various backend data sources. --- Features - πŸ“Š **Live Charts**: Bar, Line, and Gauge charts powered by Recharts. - 🧠 **Context API**: Global state management for metrics using `MetricsContext`. - 🧩 **Modular Components**: Drop-in-ready visual components for easy scaling. - 🎯 **Overview Navigation**: One-click access to the main dashboard. - βš™οΈ **Custom Hook Support**: `useSystemMetrics()` for reactive metric consumption. --- πŸ› οΈ Tech Stack | Layer | Tech | |-------------|-------------------------| | Frontend | React.js, CSS | | Charts | Recharts | | State Mgmt | React Context API | | Backend | GoLang, MongoDB | --- Project 2: name: Trip Planner Agent Workflow Built with CrewAI Plan your perfect getaway with the power of AI agents working in synergy! This intelligent workflow features specialized agents that collaborate to design a smart, data-driven, and personalized 7-day travel itinerary. Project Overview This project leverages multiple autonomous agents to plan a trip from scratch. Each agent has a clearly defined role and utilizes real-time tools to gather information and make decisions. Agents in Action Expert Travel Agent Designs a detailed 7-day travel itinerary including: Per-day plans Estimated budget Packing suggestions Safety tips Powered by GPT-4 + Tools (Search, Calculator) City Selection Expert Analyzes travel trends, weather, seasonality, pricing, and preferences to select the best city for travel. Powered by GPT-4 + Web Search Local Tour Guide Acts as your on-ground AI guide with detailed insights into the chosen city’s: Attractions Local customs Hidden gems Powered by GPT-4 + Web Search Tech Stack CrewAI β€” Multi-agent framework OpenAI GPT-3.5 & GPT-4 β€” Natural language agents SearchTools β€” To fetch real-time web data CalculatorTools β€” To calculate trip budgets Python 3.10+ Project 3: Name: GeoRisk AI Overview This project is a demonstrating a multi-agent system that leverages IBM watsonx.ai and Retrieval-Augmented Generation (RAG) to provide climate-aware risk analysis for business risk. It enables businesses to make informed decisions by assessing weather risks and historical climate data relevant to key supply chain points. Features Multi-agent architecture with specialized agents for: Location intelligence Climate data retrieval Business risk analysis Orchestrator agent to coordinate tasks IBM watsonx.ai for advanced AI-driven chat and recommendations RAG (Retrieval-Augmented Generation) to use historical climate and location data Interactive web interface: Chat panel for weather queries and supply chain risk advice Map component for selecting global locations Backend API to manage communication between frontend and agents Tech Stack Frontend: React.js Backend: Flask AI & NLP: IBM watsonx.ai Data Retrieval: RAG framework Map Integration: React map component with location picker HTTP Client: Axios project 4: Name: Ai-Conversation πŸ€– GPT vs LLaMA: The Chatbot Showdown A fun Python project where GPT-4o-mini (snarky and argumentative) and LLaMA 3 via Groq (polite and diplomatic) engage in a back-and-forth conversation. Think of it as a polite librarian chatting with an overly caffeinated debate club member. project 5: name: Webscraping using BeautifulSoup A simple yet powerful Python project that demonstrates web scraping using BeautifulSoup. This project showcases how to extract useful information from websites like quotes, news, products, or custom data β€” turning unstructured web content into structured data. Features: Scrapes data from static web pages Parses HTML content using BeautifulSoup Supports CSV/JSON export of scraped data Clean and modular codebase Customizable scraping logic Requirements: Python 3.7+ requests beautifulsoup4 Project 6: name: Mandt - E-commerce Perfume Store Welcome to Mandt, a full-stack e-commerce application designed for perfume enthusiasts! This website allows users to browse and purchase a variety of perfumes for men, women, and unisex. Built with React.js for the frontend and Node.js with Express.js for the backend, Mandt leverages MySQL for efficient database storage. Features User-Friendly Interface: Browse through an extensive collection of perfumes with ease. Product Selection: Choose from a variety of options tailored for men, women, and unisex fragrances. Secure Ordering: Place your orders securely through our streamlined checkout process. Responsive Design: Fully responsive design ensures a seamless experience on all devices. Tech Stack Frontend: React.js Backend: Node.js, Express.js Database: MySQL project 7: name:Verdicta - AI Legal Assistant Verdicta is an advanced AI-powered legal assistant designed to assist with document analysis, contract review, legal research, and compliance checking. It integrates NLP, machine learning, and deep learning techniques to provide customized legal insights. πŸš€ Features: Chatbot Interface: Ask legal questions and receive AI-generated responses. PDF & Image Upload: Extracts and analyzes text from legal documents. RAG-based Search: Retrieves relevant legal documents for context-aware answers. Emotion Recognition: Understands user emotions for better interaction. Local & Secure: Runs LLaMA 3 locally on a Lenovo Legion laptop for privacy. Tech Stack Frontend: React, Vite, React Markdown Backend: Python, Flask, LLaMA 3, ChromaDB Libraries: pdf.js, Tesseract.js, OpenAI API (Groq) Installation Prerequisites Node.js & npm Python 3.x Setup Instructions Clone the repository Install frontend dependencies Install backend dependencies Start the backend server Start the frontend server Access the app at http://localhost:3000 API Endpoints POST /query Description: Processes user questions and returns AI-generated legal responses. Request: JSON containing the user query. Response: AI-generated legal answer. Contributing We welcome contributions! Feel free to fork, submit pull requests, and open issues. License MIT License Authors Athul - RAG Specialist & Project Lead