Agentic-AI is a collection of intelligent agents and workflows built using LangGraph, LangChain, and modern LLM frameworks.
It demonstrates how AI agents can collaborate, reason, and interact with external tools for real-world applications like finance, document analysis, video summarization, and more.
- 🔗 LangGraph-powered Agents – Build modular, graph-based agent workflows.
- 💬 Conversational Chatbots – Multi-tool and memory-enabled chatbots.
- 📊 Financial AI Analyst – Agents designed for financial analysis and decision support.
- 📄 PDF Assistant – Smart assistant for document parsing and Q&A.
- 🎥 Video Summarizer – Automatic video-to-text summarization pipeline.
- 🛠️ Debugging & Tracing – Agent debugging with LangSmith integration.
- 🌦️ MCP Demos – External tool integrations like weather, math server, and more.
- AgenticLangGraph/ → Core LangGraph agent workflows & debugging
- Debugging/ → Tracing, API debugging, and graph inspection
- Financial_AI_Analyst/ → Financial data agents & analysts
- LangGraph/ → Chatbots, memory, conditional edges, tool calling, and human-in-the-loop demos
- MCP_demo/ → Modular tool-based agents (math, weather, client-server agents)
- PdfAssistant/ → Document parsing and PDF-based assistant
- VideoSummarizer/ → Agent pipeline for summarizing videos
- RecipeChefAgent_LangFlow.json → Agent design in LangFlow
- Core Frameworks: LangGraph, LangChain, LangFlow
- LLMs: OpenAI GPT, other pluggable LLMs
- Integrations: LangSmith (tracing), external APIs (weather, math), PDF parsing
- Languages: Python
- Dependencies: Managed with
pyproject.tomlandrequirements.txt
- Clone the repository:
git clone https://github.com/PRANAVBALAJIRS/Agentic-AI.git cd Agentic-AI - Create a virtual environment:
python -m venv venv source venv/bin/activate # Linux/Mac venv\Scripts\activate # Windows
- Install dependencies:
pip install -r requirements.txt
- Run a demo (example: PDF Assistant):
cd PdfAssistant python pdf_assistant.py