From the course: Building LLM-Powered Recommendation Systems
Join today to access over 25,500 courses taught by industry experts.
Production infrastructure: Vector databases and model serving
From the course: Building LLM-Powered Recommendation Systems
Production infrastructure: Vector databases and model serving
Imagine a massive, magical library. Instead of a card catalog, you can just describe the feeling of the book you want. Something like adventurous, but also thoughtful, like the Hobbit meets a philosophy text, and the librarian instantly finds that perfect match on the shelf. This is exactly what a vector database does for our data. But how does the library operate at the speed of Amazon on Netflix, serving millions of people at once without breaking a sweat? That requires a powerful, specialized operational backbone. To answer that question, we need to understand the specialized infrastructure that powers modern recommenders. In this video, we'll explore the two foundational pillars of a GenAI production system, vector databases, and dedicated model serving platforms. You will learn how these components work together to enable fast, scalable, and semantically-aware recommendations in a real-world environment. Let's start with the heart of semantic search, the vector database. Unlike a…
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.