Skip to content

AI Agent Orchestration with LangGraph, RAG & MCP | FastAPI, Docker

Notifications You must be signed in to change notification settings

Shiru99/AeroMate

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

19 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

✈️ AeroMate - AI Agent

AeroMate is an AI Agent powered by Gemini and orchestrated via LangGraph, designed to demonstrate a robust MCP (Model Context Protocol) client-server architecture. It integrates RAG, Docker-sandboxed Manim execution, and various utility tools through a FastAPI backend and Streamlit frontend to simulate a complex, production-ready agentic environment

Features

  • Gemini LLM Integration: Leverages the capabilities of Gemini for advanced language understanding and generation.
  • LangGraph Orchestration: Utilizes LangGraph to manage and coordinate multiple models
  • RAG (Retrieval-Augmented Generation): Implements RAG techniques to enhance response accuracy and relevance.
  • MCP Client-Server Architecture: Demonstrates a scalable and modular architecture for AI
  • Docker-Sandboxed Manim Execution: Executes Manim animations in a secure Docker environment.
  • FastAPI Backend: Provides a robust backend for handling requests and managing the AI agent
  • Streamlit Frontend: Offers an interactive and user-friendly interface for users to interact with the AI agent.

Test Prompts

  • Create a Matrix-inspired intro. Display a chaotic cloud of green binary digits (0s and 1s) rapidly changing in the center of the screen. Suddenly, snap these chaotic numbers into a grid structure that spells out 'AEROMATE'. Then, flash the text white and stabilize it. Add the subtext 'AI Agent by Shiru99' at the bottom.

  • Hello, I’m Shiru99. Tell me 2 interesting facts about the Google Company

  • I would like to know it's stock price, in USD as well as in INR. If I invest INR 1,78,912 today in this stock, which has an expected 12.5% IRR for the next two years, at the same time, USD is expected to have an annual appreciation of 5%, what would be my overall return after 2 years in INR? Give me the formula used or calculations.

  • I was involuntarily downgraded on a flight from Delhi to New York, what is the specific reimbursement am I entitled to?

  • Create a Matrix-themed outro animation with neon-green Monospace text on a black background. First, display 'Thank You!' in the center. Next, smoothly transform it into a small header at the top that says 'Powered By:', and below it, create two separate vertical columns of text: the left column containing 'LangChain', 'Gemini LLM', and 'RAG & MCP', and the right column containing 'Custom Tools', 'FastAPI', and 'Docker'. Ensure these two columns are positioned with a significant horizontal gap between them so they do not overlap, and scale the entire group down to fit within the center of the screen. Hold this for 2 seconds. Finally, transform the entire scene into a small, subtle, and elegant white font positioned at the bottom center of the screen that reads 'Built by Shiru99', ensuring this final line is much smaller than the previous text for a humble finish

About

AI Agent Orchestration with LangGraph, RAG & MCP | FastAPI, Docker

Resources

Stars

Watchers

Forks