Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/jbarrasa/goingmeta/llms.txt

Use this file to discover all available pages before exploring further.

Every Going Meta session ships with a dedicated folder in the GitHub repository, a YouTube recording, and a row in the README session table. Understanding how these three pieces fit together makes it much easier to follow along, reproduce results, and adapt the code for your own projects. This guide explains the conventions used across all 46+ sessions.

​
Repository Structure

The repository is organised as a flat list of session folders at the root level, named session01 through session46 (and growing). Each folder is self-contained — you only need to clone or download the folder relevant to the session you’re studying. 
goingmeta/
├── README.md               ← Master session table (all seasons)
├── session01/              ← Season 1, Episode 1
│   ├── cypher/
│   └── resources/
├── session02/              ← Season 1, Episode 2
│   ├── build-kg.cypher
│   └── resources/
│       ├── devto-articles.csv
│       └── goingmeta-skos.ttl
├── session05/
│   └── Ontology_Driven_KG_creation.ipynb
├── session27/
│   └── langgraph_reflection_agent.py
├── session32/
│   ├── pyproject.toml
│   └── app.py
├── session34/
│   └── python/
│       └── dynamic.py
└── ...
Season numbers do not reset the folder numbering. Season 2 begins at session28, Season 3 at session39. This makes it easy to jump to any session by its global episode number shown in the README table.

​
File Types You Will Encounter

Different sessions use different artefact types depending on the topic. Here’s what each type represents:

Cypher Scripts (.cypher)

Direct Neo4j queries — create constraints, import RDF, build graph models, run SPARQL via n10s. Run these in Neo4j Browser or via cypher-shell.

Jupyter Notebooks (.ipynb)

Interactive Python notebooks, often runnable in Google Colab with a single click. Cover ontology processing, ETL pipelines, and LLM integration experiments.

Python Scripts (.py)

Standalone scripts for agents, GraphRAG pipelines, and Streamlit apps. Usually read configuration from environment variables (see Setup).

Ontology Files (.ttl)

Turtle-serialised RDF ontologies and SKOS vocabularies that can be imported directly into Neo4j using Neosemantics (n10s).

​
Reading the Session Table

The README contains one table per season. Each row has six columns:
ColumnMeaning
#Session number within the season (global number used for folder naming)
broadcastOriginal air date (month and day)
titleFull episode title describing the main topic
tagsTechnology and concept tags (e.g. LLM, Ontology, SPARQL, Python)
recording📺 link to the YouTube/Twitch replay
code💻 link to the session folder in this repository
Use the tags column to find sessions by technology. For example, filter for GraphRAG to find all episodes in the ontology-driven RAG arc, or SHACL to find data-quality episodes.

​
How Sessions Are Organised by Season

Sessions 1–27, broadcast from February 2022 through April 2024. Covers semantic-web foundations: RDF integration patterns, SPARQL, SHACL, ontology reasoning, graph algorithms, Wikidata/DBpedia integrations, and early LLM experiments with OpenAI. Most artefacts are Cypher scripts and Jupyter notebooks.

​
Tips for Getting the Most Value

1

Watch first, then run the code

Each YouTube recording walks through the concepts and live-codes the session. Watching the recording first gives you the mental model before you try to run the scripts yourself.
2

Set up your environment once

Most sessions share the same Neo4j + Python stack. Follow the Setup guide once and you will be ready for the vast majority of sessions without per-session configuration.
3

Use tags to build a learning path

If you want to learn GraphRAG end-to-end, filter the README for GraphRAG tags: sessions 22, 23, 24, 31, 32, 33, 34. Each builds on the previous. Likewise, Ontology + Python gives you the KG-construction arc from sessions 5, 25, 28, 29, 30.
4

Open notebooks in Colab

Jupyter notebooks in the repository include a Colab badge at the top. Click it to open an interactive environment with zero local setup — ideal for experimenting with ontology processing and NLP pipelines.
5

Adapt the Cypher scripts

Cypher scripts in each session folder are written to be readable and instructive. Copy them into Neo4j Browser and modify the URIs or labels to work with your own data.
The README session table is the fastest index into the entire series. Use your browser’s Ctrl+F / Cmd+F to search for a technology name (e.g. LangGraph, SHACL, Wikidata) and jump directly to the sessions that cover it.

​
Where to Ask Questions

The best place to ask questions is in the comments section of each YouTube video. Jesus Barrasa monitors them regularly. You can also open issues or discussions in the GitHub repository for code-specific questions.
All code in the repository is open source. Check the repository’s licence file for the exact terms. The code is intended to be educational and is designed to be adapted for production use cases.
Most sessions use the current stable Neo4j release at the time of broadcast. For sessions 1–20, Neo4j 4.x was current; from session 21 onward, Neo4j 5.x patterns are used. Check the session README or the recording intro for the exact version if it matters for your environment.

Build docs developers (and LLMs) love

Get started for freeTalk to us