Building GenAI-Powered Apps with MariaDB’s VectorDB and SkySQL’s AI Agents
Abstract
In this session, we’ll explore how MariaDB’s VectorDB along with SkySQL’s SkyAI Agents transform your database into a foundation for advanced AI applications by enabling semantic search, “vector embedding” management, and conversational AI workflows. You’ll learn how SkyAI Agents offer an opinionated, in-database solution for Natural language conversations incorporating fully managed RAG pipelines, that prioritizes accuracy, reliability, and developer simplicity.
Description
MariaDB has rolled out a significant enhancement to its server: MariaDB Vector, a built-in vector search capability. This presentation will introduce you to the core functionality of MariaDB Vector, explaining how it works, when to use it, and when you might opt for a different solution. We’ll compare its performance and features to other vector databases, highlight areas where it still has room to grow, and discuss the future roadmap. By attending, you’ll be fully equipped to begin integrating MariaDB Vector into your AI projects—or to make an informed decision if it’s not the right fit.
Description – We will walk through our high level architecture for how we integrated the Agentic RAG pipeline with MariaDB. We will demo:
- Training SkyAI Agents by combining database metadata, data insights, and human-in-the-loop semantics
- Show how the agents use Vector Data to manage categorical values enabling semantic searches.
- Deploy and integrate your agents using simple REST APIs to deliver task-specific AI that reasons, decomposes complex queries, and auto-corrects SQL errors.
By the end of this session, you’ll see how MariaDB’s VectorDB, paired with SkyAI Agents, removes the friction from building intelligent applications, enabling you to focus on solving real-world problems with AI-powered precision.