This repository archives code and resources from the "Going Meta" live stream series, focusing on knowledge graph (KG) construction, semantic technologies, and their integration with Large Language Models (LLMs). It targets developers, data scientists, and researchers interested in practical applications of graph databases, ontologies, RDF, SPARQL, and LLM-powered graph solutions. The primary benefit is access to practical examples and code for building and leveraging KGs.

How It Works

The project provides a collection of code examples and resources demonstrating various techniques for knowledge graph manipulation and LLM integration. It covers topics such as semantic search, ontology-driven reasoning, data quality with SHACL, RDF integration, and advanced Retrieval-Augmented Generation (RAG) patterns using tools like LangChain, LangGraph, and Neo4j. The approach emphasizes practical, hands-on demonstrations of these technologies.

Quick Start & Requirements

• Install/Run: Specific commands vary per session's code examples. Many examples utilize Python libraries and may require Neo4j or other graph databases.
• Prerequisites: Python, Neo4j (AuraDB or local), RDFLib, LangChain, potentially specific LLM API keys, and various data science libraries.
• Resources: Links to recordings and code are provided for each session.

Highlighted Details

• Comprehensive coverage of KG and LLM integration, from basic semantic search to advanced RAG patterns.
• Demonstrations of integrating ontologies for data quality, reasoning, and guiding KG construction.
• Practical examples using Python, Neo4j, RDF, SPARQL, SHACL, and LLM frameworks like LangChain.
• Sessions span from 2022 to ongoing 2025, indicating a sustained focus and evolving content.

Maintenance & Community

The project is associated with live streams held monthly, suggesting an active, albeit community-driven, development and content creation process. Specific maintainer details or community links (e.g., Discord/Slack) are not explicitly provided in the README.

Licensing & Compatibility

The repository itself does not specify a license. The code examples within may be subject to the licenses of the individual libraries used (e.g., Python libraries, Neo4j). Users should verify licensing for any code or data they intend to use commercially.

Limitations & Caveats

This repository serves as an archive of past live stream content; it is not a cohesive library or framework. Users will need to navigate individual session folders to find relevant code and dependencies, which may vary significantly. The project's structure is a collection of disparate examples rather than a unified tool.