Skip to content

divakarkumarp/Building-Agentic-AI

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

12 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Building-Agentic-AI

Agentic AI 🤖

Python Jupyter OpenAI Langchain AutoGen Phidata CrewAI PraisonAI Pydantic

License: MIT GitHub Stars GitHub Forks GitHub Issues

A comprehensive collection of different AI agent implementations using various frameworks and approaches.

Project Overview

This repository contains implementations of different AI agents using popular frameworks like AutoGen, CrewAI, LangGraph, and more. Each implementation demonstrates different approaches to building autonomous AI agents for various use cases.

Agent Implementations

Agent Framework Location Description
Agno (Phidata) /Agno(phidata) Advanced implementations with DeepSeek UI integration for RAG, financial analysis agents, web search and research tools, playground environment for testing, and database-integrated agents.
AutoGen /Auto_Gen Multi-agent implementation using AutoGen framework demonstrated through Python scripts and interactive Jupyter notebooks.
CrewAI /CrewAI Collaborative AI agents featuring ESG applications, API integration examples, interactive Jupyter notebooks for testing, and user input handling implementations.
LangGraph /Langgraph_Agent Extensive collection of agents including ReAct pattern implementations, database interaction, structured reports, tool-augmented agents, RAG implementations, and practice notebooks.
LangGraph Special Agents /Langgraph_Agent/LangGraph_1 Specialized implementations including virtual insurance agents (Advisr), custom support chatbots, human-in-loop agents, multi-agent systems, spreadsheet AI agents, and vision-enabled agents.
PraisonAI Agents /PraisonAI_agents_mcp Communication platform integrations including Airbnb search functionality, WhatsApp integration with multi-agent support, Slack integration capabilities, and WhatsApp bridge with MCP server.
Pydantic AI /Pydantic_ai Type-safe AI applications with Pydantic model implementations, UI application examples, and testing implementations.

Project Structure

Core Components

  • Multi-agent Systems

    • AutoGen-based implementations
    • CrewAI collaborative agents
    • LangGraph multi-agent frameworks
  • RAG Systems

    • DeepSeek UI integration
    • Vector database implementations
    • Multiple data source handling
  • Specialized Agents

    • Financial analysis
    • Customer support
    • Insurance advisory
    • Communication platform integration

Implementation Details

Data Processing

  • SQL database integration
  • PDF document handling
  • CSV processing
  • SQLite database usage
  • Vector stores

Agent Capabilities

  • Tool augmented agents
  • Web search and analysis
  • Human-in-loop systems
  • Structured output generation
  • Vision processing
  • Natural language understanding

Development Tools

  • Jupyter notebooks for interactive development
  • Python scripts for production deployment
  • UI implementations
  • Testing frameworks

Requirements

Each agent implementation has its own requirements.txt file in its respective directory. Please refer to the specific requirements file for each implementation you want to use.

Getting Started

  1. Clone the repository

    git clone https://github.com/divakarkumarp/Building-Agentic-AI.git
  2. Navigate to the specific agent implementation directory

    cd Building-Agentic-AI/[agent-directory]
  3. Install the required dependencies

    pip install -r requirements.txt
  4. Follow the implementation-specific README or notebook instructions for detailed setup and usage

Project Documentation

Each major component includes:

  • README files with specific setup instructions
  • Jupyter notebooks with examples and documentation
  • Python scripts with implementation details
  • Requirements files for dependency management

About

Different AI agent implementations using various frameworks and approaches

Topics

Resources

Stars

Watchers

Forks

Packages

No packages published