From the course: Build with AI: Agentic Applications with LlamaIndex and MCP

Unlock this course with a free trial

Join today to access over 25,200 courses taught by industry experts.

Why we use vector databases to store knowledge

Why we use vector databases to store knowledge - LlamaIndex Tutorial

From the course: Build with AI: Agentic Applications with LlamaIndex and MCP

Start my 1-month free trial Buy for my team

Why we use vector databases to store knowledge

“

Thanks to Anna for that introduction to agents and agentic workflows. Now, let's switch gears a little and talk about vector databases. Generative AI models are amazing, but they have a key shortcoming. Their knowledge can be incomplete or become out of date over time. Vector databases can address this by giving generative AI models and agents instant access to current domain-specific information. To make this concrete, let's consider a practical example that we'll use throughout this chapter, an e-commerce shopping assistant. This kind of agent needs to know about the product catalog, including descriptions, prices, and inventory. But an agent's knowledge needs don't stop there. In the real world, a customer support bot must know the company's latest product specifications and policies. A personal assistant needs access to your schedules, e-mails, and notes, and a research agent should refer to the latest evidence. Unfortunately, this information is either unavailable or outdated if…

Contents