How combining graph and vector search transforms context and reduces hallucinations
What if your LLM could answer complex questions more accurately by reasoning through a knowledge map, not just reading flat text?
In this video, Alessandro Pireno, Solutions Director at SurrealDB walks through exactly how to make that happen by combining knowledge graphs with semantic search. He demos techniques for extending context windows, building smarter retrieval pipelines, and constructing graphs from even unstructured data.
Tune in to explore practical methods for building and integrating graph-powered RAG systems using open tools, compact corpora, and some clever prompting.
The talk was recorded as part of a recent Stack Session, our free professional webinar series series for AI/ML professionals.
Presentation Highlights
This talk is for engineers, data scientists, and ML practitioners building RAG systems or experimenting with hybrid search:
- Why semantic search alone can miss key relationships — and how graphs fill in the gaps.
- How knowledge graphs extend an LLM’s reasoning and reduce hallucinations by encoding relationships
- Techniques to generate graphs from unstructured data using LLMs, NLP libraries, or full-text search
- Tradeoffs across different graph generation methods: cost, complexity, and accuracy
- How SurrealDB supports multimodal (graph + vector) search in a single, Rust-built platform
About the Speaker
Alessandro (“Sandro”) Pireno is a seasoned data scientist and self-proclaimed data nerd with extensive experience in data infrastructure and graph-based technologies.
Currently working as Director of Solutions at SurrealDB, a cutting-edge database company, he explores innovative ways to make AI systems more intelligent and context-aware through knowledge graphs and vector databases.
3 Days of Context, Insights, and Connections
The 6th annual MLOps World | GenAI Summit is taking place October 7–9, 2025 at the Austin Renaissance Hotel.
Don’t miss this chance to accelerate and de-risk your AI/ML, agentic, and infrastructure outcomes through cutting-edge strategies, real-world case studies, technical deep dives, and hands-on workshops. Every presentation is hand-picked by a committee of fellow AI practitioners to spotlight where the line of excellence is, right now.
The experience includes a vibrant expo, where attendees shift from focused learners to active participants by engaging in Brain Dates, Community Stage, Startup Zone, interactive demos with top vendors.
MLOps World | GenAI Summit is a high-impact way to learn, connect, and elevate your team, projects, and career.
Early Bird tickets are on sale now and offer a 15% savings when you register in advance.