Melbourne Data Eng Meetup, Graph Edition

Hey all, please join us for our November Edition of the Melbourne Data Eng Meetup which will dive deep on all things Knowledge Graphs and GraphRAG. Join us for an engaging session filled with exciting discussions and networking opportunities.

Don't miss out—RSVP to secure your spot. Please reach out if you'd like to host us! 🎉

🏠 Location: TBC
🍕 Catering: Neo4j 😊

Schedule:
- 5.30pm: Doors Open
- 6.00pm: Announcements and Welcome
- 6.10pm: 1st Talk
- 6.40pm: 2nd Talk
- 7.10pm: Networking
- 7.30pm: Doors Close

Speakers:
🎤 Fanghua (Joshua) Yu, Founder, GraphWay
Talk Title: Has RAG Kept Its Promises? Hallucination, factuality, explainability—wins vs gaps
Talk Summary: This is a clear, practical look at RAG in 2025. We’ll talk about what really cuts mistakes, what still goes wrong, and simple ways to make RAG fast, cheap, and reliable. Topics include: better chunking and retrieval, when re-ranking helps, when a knowledge graph is worth it, and where agents or images/tables fit.
Speaker Bio: Fanghua (Joshua) Yu is a graph-first AI builder focused on turning structured data and unstructured documents into trustworthy answers. He leads GraphWay AI, developing document parsing, knowledge extraction, and GraphRAG pipelines that run efficiently on modest GPUs.Joshua writes frequently on Medium, covering RAG architectures, knowledge graphs, and multimodal retrieval, and speaks to practitioners about making GenAI production-ready.With a background in solution design and data platforms, he blends knowledge graphs with LLMs to reduce hallucinations, improve traceability, and drive outcomes for enterprise analytics. He enjoys connecting communities and sharing playbooks for fast, affordable, explainable RAG at real scale.
🎤 Jonathan Carroll, Principal Consultant / Data Scientist / Engineer / Developer, Irregularly Scheduled Programming
Talk Title: Graph Database Projects for Fun and Profit
Talk Summary: Is a Graph Database a good fit for the shape of this particular data? Sometimes it is, sometimes it isn't. In this talk I'll demonstrate three projects I've worked on involving Neo4j databases - one for fun involving an actual "knowledge" graph extracted from my Obsidian notes; one for profit involving identifying opportunities for drug repurposing; and one you might not expect involving maths puzzles.What makes Neo4j a good fit in each of these cases? What does it offer beyond more traditional relational data modelling? What are the challenges involved with getting data in and out of a graph database that need to be considered? I'll aim to answer all of these and hopefully inspire you to add a graph database to your next project.
Speaker Bio: Jonathan Carroll is an independent contractor and principal consultant at Irregularly Scheduled Programming, currently remotely contracting with biotech companies in USA and Europe. He received a PhD in theoretical physics from the University of Adelaide where he programmed mainly in Fortran. He has worked in the South Australian public service in fisheries, programming in R. He has since branched out to polyglot programming in Python, Julia, Rust, Haskell, APL, and a variety of other languages. Jonathan is an editor of the RWeekly (rweekly.org) newsletter, regularly blogs about explorations in programming at jcarroll.com.au and is active on mastodon at fosstodon.org/@jonocarroll

Talk details will be available soon. Interested to speak at this or future meetups? Fill this form: https://dev.neo4j.com/submit-your-talk

Whether you're a graph geek, a cloud-native builder, or an AI tinkerer, come be a part of this conversation. See you there!