GraphRAG - Connecting the Dots for Information Discovery

The classic RAG application extends an LLM with private information, able to fetch answers to questions that are contained in a single chunk of text. What if the answer requires connecting the dots across multiple chunks that may not be directly similar to the question? That is information discovery with GraphRAG. 

You'll learn how to: 

- reconstruct chunks into the original context

 - meaningfully connect disparate chunks 

- expand unstructured text data with structured data 

- combine all this into a RAG workflow


Andreas Kollegger

Software Engineer @Neo4J

Andreas is a technological humanist. Starting at NASA, Andreas designed systems from scratch to support science missions. Then in Zambia, he built medical informatics systems to apply technology for social good. 

Now with Neo4j, he is democratizing graph databases to validate and extend our intuitions about how the world works. Everything is connected.

Read more
Find Andreas Kollegger at:

Session Sponsored By

Neo4j, the Graph Database & Analytics Company, helps find hidden data relationships & patterns.


Wednesday Apr 10 / 11:45AM BST ( 50 minutes )


Westminster (4th Fl.)


Video is not available


From the same track


Simplifying Streaming Data Transforms with WebAssembly

Wednesday Apr 10 / 10:35AM BST

The majority of stream processing work is spent on mundane transformation tasks like scrubbing, normalisation, and filtering. Yet to perform these relatively simple tasks involves standing up multiple distributed systems that add complexity and take time to learn.

Speaker image - James Kinley
James Kinley

Principal Solutions Architect @Redpanda

Speaker image - Paul Wilkinson
Paul Wilkinson

Principal Solutions Architect @Redpanda


Elevating Developer Experiences with AI-Powered API Testing & Documentation

Wednesday Apr 10 / 01:35PM BST

In the modern era, APIs have exploded in their popularity, powering absolutely everything we interact with on a daily basis (whether we know it or not). Maintaining their coherence and functionality has become, to say the least, a difficult task for developers and engineering leaders.

Speaker image - Jamie White
Jamie White

Strategic Solutions Engineer @Postman


Building InfluxDB 3.0 using Modular Open Source Components

Wednesday Apr 10 / 03:55PM BST

In this talk, we will describe the design of InfluxDB 3.0's new high performance time series database engine using Apache Arrow, DataFusion, Parquet and Rust.

Speaker image -  Raphael  Taylor-Davies
Raphael Taylor-Davies

Senior software engineer @InfluxData


An Introduction to Event APIs

Wednesday Apr 10 / 02:45PM BST

Learn how asynchronous communication can revolutionize business, positively impact scalability and real-time decision-making, and grasp synchronous-asynchronous federation technologies for seamless integration.

Speaker image - Linus Hakansson
Linus Hakansson

Field CTO @Gravitee