Agentic AI - Private Agentic RAG with LangGraph and Ollama

<div>This course is not for absolute beginners in AI - you should first learn LangChain fundamentals, then LangGraph, and only after that take this course for the best learning experience.</div><div><br></div><div>Private Agentic RAG with LangGraph and Ollama is an advanced, project-based course that teaches you how to build private, production-ready Retrieval-Augmented Generation (RAG) systems using LangGraph, LangChain, Ollama, ChromaDB, Docling, and Python.</div><div><br></div><div><span style="font-size: 1rem;">This course is designed for developers who want strong control over their data, full privacy, and complete end-to-end workflows using local LLMs.</span></div><div><br></div><div><span style="font-size: 1rem;">You will learn how to build modern RAG systems, implement advanced retrieval pipelines, add agent workflows, use LangGraph state machines, integrate SQL agents, and run everything on your own machine using Ollama. All projects run 100 percent locally, with no external API cost and no data leaving your system.</span></div><div><br></div><div>The entire course is practical. Every concept is explained with step-by-step notebooks, complete Python code, and real examples using SEC financial filings from Amazon, Google, Apple, and Microsoft.</div><div><br></div><div>What You Will Learn</div><div><span style="font-size: 1rem;">Ollama and Local LLM Setup</span><br><ul><li><span style="font-size: 1rem;">Install and configure Ollama for private LLM deployment</span></li><li><span style="font-size: 1rem;">Use models like Qwen3, GPT-OSS, Llama 3.2, and nomic-embed</span></li><li><span style="font-size: 1rem;">Create custom LLMs with Modelfiles</span></li><li><span style="font-size: 1rem;">Use Ollama CLI and REST API for text, chat, and embeddings</span></li></ul></div><div><span style="font-size: 1rem;">LangGraph Fundamentals</span></div><div><ul><li>Build state machines using TypedDict</li><li><span style="font-size: 1rem;">Create nodes, reducers, and conditional edges</span></li><li><span style="font-size: 1rem;">Build multi-step workflows with START/END logic</span></li><li><span style="font-size: 1rem;">Visualize execution with diagrams</span></li><li><span style="font-size: 1rem;">Understand message accumulation and state merging</span></li></ul></div><div><span style="font-size: 1rem;">Complete RAG Systems (from scratch)</span></div><div><ul><li><span style="font-size: 1rem;">Ingest PDFs using Docling with OCR and table extraction</span></li><li><span style="font-size: 1rem;">Build page-level chunks for accurate retrieval</span></li><li><span style="font-size: 1rem;">Extract metadata from filenames and LLMs</span></li><li><span style="font-size: 1rem;">Remove duplicates using SHA-256 hashing</span></li><li><span style="font-size: 1rem;">Store documents in ChromaDB with metadata filters</span></li></ul></div><div><span style="font-size: 1rem;">Two-Stage Retrieval Pipeline</span></div><div><ul><li>Build metadata filters from natural language</li><li><span style="font-size: 1rem;">Generate financial keywords using structured LLM outputs</span></li><li><span style="font-size: 1rem;">Use ChromaDB with MMR search</span></li><li><span style="font-size: 1rem;">Implement BM25Plus re-ranking for better accuracy</span></li><li><span style="font-size: 1rem;">Extract headings and sections for improved ranking</span></li></ul></div><div><span style="font-size: 1rem;">Agentic RAG using LangGraph</span></div><div><ul><li>Build tool-calling agents using the ReAct pattern</li><li><span style="font-size: 1rem;">Implement document retrieval tools using LangChain</span></li><li><span style="font-size: 1rem;">Build agents that call tools multiple times</span></li><li><span style="font-size: 1rem;">Add table-based answers with citations</span></li><li><span style="font-size: 1rem;">Support multi-turn conversations with memory</span></li></ul></div><div><span style="font-size: 1rem;">Corrective RAG (CRAG)</span></div><div><ul><li>Grade retrieved documents using a Pydantic schema</li><li><span style="font-size: 1rem;">Detect irrelevant results and rewrite queries</span></li><li><span style="font-size: 1rem;">Add web search fallback using DuckDuckGo</span></li><li><span style="font-size: 1rem;">Prevent infinite loops with controlled retries</span></li><li><span style="font-size: 1rem;">Generate final answers with correct citations</span></li></ul></div><div><span style="font-size: 1rem;">MySQL SQL Agent</span></div><div><ul><li><span style="font-size: 1rem;">Build a natural-language SQL agent with LangGraph</span></li><li><span style="font-size: 1rem;">Retrieve schema, generate SQL, validate, run, and fix errors</span></li><li><span style="font-size: 1rem;">Handle multi-table joins and complex metrics</span></li><li><span style="font-size: 1rem;">Automatically correct broken SQL queries</span></li><li><span style="font-size: 1rem;">Support explanations and safe database access</span></li></ul></div><div><span style="font-size: 1rem;">Financial Document Analysis Project</span></div><div><ul><li>Work with real SEC filings: 10-K, 10-Q, 8-K</li><li><span style="font-size: 1rem;">Build a complete RAG system that answers questions like:</span></li><li><span style="font-size: 1rem;">“What was Amazon’s revenue in 2023?”</span></li><li><span style="font-size: 1rem;">“Compare Google and Apple’s cash flow for 2024”</span></li><li><span style="font-size: 1rem;">“Show segment revenue with citations and tables”</span></li><li><span style="font-size: 1rem;">Use ChromaDB + BM25 for accurate retrieval</span></li><li><span style="font-size: 1rem;">Produce clean, formatted answers with tables and reasoning</span></li></ul></div><div><span style="font-size: 1rem;">Who This Course Is For</span></div><div><ul><li><span style="font-size: 1rem;">Developers and engineers who want to build advanced RAG systems</span></li><li><span style="font-size: 1rem;">ML practitioners who want full privacy using local LLMs</span></li><li><span style="font-size: 1rem;">AI engineers working on LangGraph, LangChain, or agent systems</span></li><li><span style="font-size: 1rem;">Backend developers who want to build real GenAI applications</span></li><li><span style="font-size: 1rem;">Anyone interested in private, production-grade LLM workflows</span></li></ul></div><div><span style="font-size: 1rem;">This is an advanced-level course. Good LangGraph or Langchain knowledge is required.</span></div><div><br></div><div>Why This Course Is Different</div><div><ul><li>The entire course runs locally using Ollama</li><li><span style="font-size: 1rem;">Zero API cost and complete data privacy</span></li><li><span style="font-size: 1rem;">Covers modern RAG techniques: PageRAG, CRAG, Reflexion ideas</span></li><li><span style="font-size: 1rem;">Real datasets from top tech companies</span></li><li><span style="font-size: 1rem;">Covers LangGraph deeply with real production workflows</span></li><li><span style="font-size: 1rem;">Includes SQL agents, financial RAG systems, and multi-step agents</span></li><li><span style="font-size: 1rem;">Step-by-step, practical, and code-heavy</span></li></ul></div><div><span style="font-size: 1rem;">By the End of This Course You Will Be Able To</span></div><div><ul><li>Build private, production-ready RAG systems</li><li><span style="font-size: 1rem;">Deploy and fine-tune local LLMs with Ollama</span></li><li><span style="font-size: 1rem;">Build graph-based agents using LangGraph v1</span></li><li><span style="font-size: 1rem;">Create advanced retrieval pipelines using MMR and BM25Plus</span></li><li><span style="font-size: 1rem;">Analyze financial documents with precise citations</span></li><li><span style="font-size: 1rem;">Build SQL agents for natural language database queries</span></li><li><span style="font-size: 1rem;">Handle query rewriting, grading, and web fallback</span></li><li><span style="font-size: 1rem;">Build complete agentic RAG applications end-to-end</span></li></ul></div>

What you'll learn:

• Build private, production-ready Agentic RAG systems using LangGraph v1 and Ollama.
• Create custom LLM workflows with LangGraph state machines, nodes, edges, and conditional routing.
• Implement PageRAG, metadata extraction, PDF processing with Docling, and page-level ingestion.
• Use ChromaDB, embeddings, metadata filtering, and MMR retrieval for high-accuracy search.
• Apply BM25+ re-ranking and advanced retrieval pipelines for financial document analysis.
• Build Agentic RAG: tool calling, reasoning loops, structured outputs, and multi-step workflows.
• Implement Corrective RAG (CRAG) with document grading, query rewriting, and web search fallback.
• Create custom Ollama models, Modelfiles, embeddings, and integrate with LangChain.
• Build Reflexion, Self-RAG and Adaptive RAG along with MySQL Agent