Skip to content

JasonHaley/memphis-agentcamp-2026

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

11 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Memphis AgentCamp 2026 - Agentic RAG Demos

Demo scripts for the Memphis AgentCamp Agentic RAG presentation, walking through a hands-on progression from document ingestion to fully agentic RAG pipelines using Azure OpenAI and the agent-framework SDK.

Demos

The demos are numbered to follow the presentation flow, each building on concepts from the previous step.

# Demo Description
1 Ingestion Phase Chunks markdown documents, generates embeddings via Azure OpenAI, and uploads them to an Azure AI Search index.
2 Semantic Search Runs vector-only and hybrid search queries against the index, comparing how different search modes rank results.
3 Simple RAG A basic retrieval-augmented generation pipeline that retrieves context chunks and generates answers with source citations.
4 Agent Framework Hello World Minimal example using the agent-framework library with Azure OpenAI and Azure AD authentication.
5 Agentic RAG An agent that autonomously calls multiple tools (knowledge base search, support tickets, Bing grounding) to answer questions via multi-hop reasoning.
6 Agentic RAG with Knowledge Base Uses AzureAISearchContextProvider in agentic mode, letting the agent plan and execute multi-hop queries across a knowledge base with multi-turn conversation context.

Utilities

Utility Description
pdf_to_markdown.py Lightweight PDF-to-Markdown converter using the markitdown library.
pdf_to_markdown-manual.py More sophisticated converter using pdfplumber/pypdf with heuristic heading detection and table extraction.
index_loader.py Reads support ticket data from CSV, creates an Azure AI Search index with dual embeddings, and uploads documents for the agentic RAG demo.

Getting Started

# Install dependencies (requires uv and Python 3.13)
uv sync

# Run any demo
uv run python demos/01-ingestion-phase.py

Requires a .env file with Azure OpenAI and Azure AI Search credentials. See CLAUDE.md for the full list of required environment variables.

NOTE: The sample uses a Kaggle dataset from Tobias Bueck. (2025). Customer IT Support - Ticket Dataset. I have filtered out the non-english tickets and added a random create date in order to perform additional search types.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages