Proof-of-concept examples for agentic patterns, primarily Microsoft AutoGen, plus general LangChain-agent and prompt-engineering notebooks that don't belong to any one framework's dedicated folder.
conda env create -f environment.yml
conda activate agentsMulti-agent conversation examples using Microsoft AutoGen — function calling, group chat (with and without visualization), math chat, two-player setups, and feedback from code execution.
langchain_camel_agent.py— CAMEL (role-playing) multi-agent conversationlangchain_transformer_agents.py— HuggingFace Transformers + LangChain agent
Standalone prompt-engineering notebooks (ChatGPT, Gemini, Mixtral, Code Llama, RAG, ReAct, PAL, function calling, litellm) — not agent-framework-specific.
agents_A2A-MCP-Tutorial_PavanBelagatii.ipynb— Agent2Agent/MCP protocol tutorialManim-Coder-FineTuning-Experiment.ipynb— unrelated fine-tuning experimentconvert-llama-ggml-to-gguf.py,download_llama2.py— model utility scripts
CrewAI and LangGraph content has been consolidated into their own dedicated top-level
directories (Code/crewai/, Code/langgraph/) — including the CrewAI-in-LangGraph
hybrid examples, which live in Code/langgraph/ since LangGraph is the driving
framework there. This folder is now the AutoGen home plus cross-framework/general PoCs.