Text-to-query agents

1. Text-to-query agents

Hi, and welcome to the course!

2. Meet your instructor...

I'm Apoorva Joshi, a Senior AI Developer Advocate at MongoDB, and I'll be your instructor for this course.

3. Coming up...

Together, we'll learn how to create an AI agent that can query a database to answer questions from your data. These are often called text-to-query agents. We'll be using MongoDB, a world-class document database designed with applications, including AI applications, in mind. However, the concepts and approaches discussed in this course can apply to any other database provider and querying language.

4. RAG and Agents

To achieve our goal of a text-to-query agent, we'll be combining two important concepts in AI engineering: Retrieval Augmented Generation, or RAG, and AI agents.

5. RAG and Agents

RAG is the process of retrieving information from external data sources, augmenting a prompt with the retrieved information, and passing it to an LLM so it can generate more accurate answers to questions on topics it wasn't trained on. AI agents use LLMs to make decisions and trigger actions through the use of tools. RAG can be used as a tool for an AI agent, meaning an agent can decide whether it needs to access an external data source to answer a question, and if it does, trigger one or a series of tool calls to retrieve information. In our case, this will involve querying a database using MongoDB's Query API.

6. MongoDB Atlas

Our first step on this journey is getting setup with a free MongoDB Atlas account and creating our own cluster. MongoDB Atlas is a fully-managed cloud database service designed to simplify deploying, scaling, and managing MongoDB databases.

7. MongoDB Atlas

A cluster, in this case, can be thought of as the machinery that supports and runs the database that houses the data. This cluster comes with a pre-built movies database, and we'll build a text-to-query agent that uses an LLM and a set of tools to query the database and answer questions about it.

8. Course format

Throughout the course, we'll be working side-by-side to tackle exercises, which will be completed in DataLab Notebooks with accompanying videos. DataLab is DataCamp's proprietary cloud-based notebook IDE. If you've used other notebook IDEs, you should find DataLab pretty intuitive; if you're not familiar with notebooks, you can think of a notebook as a series of cells to execute your code in. If you run one cell, any variables or libraries defined in that cell will be available in subsequent cells, and so on. Working in these notebooks will give you greater insight into how to build text-to-query agents for your own projects. In the exercise, you can either start the video and code-along with us, or you can watch the video first and code afterwards - the choice is yours!

9. Course format

You can find detailed information to help you complete the tasks in the instructions section, and in the notebook.

10. Let's practice!

We'll discuss the text-to-query workflow in more detail in the next video, but for now, let's get set up with our MongoDB cluster and begin connecting to it.

Create Your Free Account

By continuing, you accept our Terms of Use, our Privacy Policy and that your data is stored in the USA.