Building Multimodal AI Applications Using MongoDB and Voyage AI

Session Pre-Requisites

• MongoDB cluster setup:
  • Register for a free MongoDB Atlas account.
  • Create a new database cluster
  • Obtain the connection string for your database cluster
• Obtain a Voyage AI API key: Follow the steps here to get a Voyage AI API key.
• Obtain a Gemini API key: Follow the steps here to get a Gemini API key via Google AI Studio.

As AI applications expand beyond text, the ability to work with image, video, and other modalities is becoming a must-have skill. Building effective multimodal systems requires not only the right models, but also the right infrastructure to store, retrieve, and serve diverse data types at scale.

In this code-along, Apoorva Joshi, a Senior AI Developer Advocate at MongoDB, will teach you how to build a simple multimodal AI application using MongoDB and Voyage AI. You’ll learn how to structure and query image and video data, apply retrieval techniques with Voyage AI, and connect everything in a functional pipeline. This session is ideal for data scientists and AI engineers looking to expand their application-building toolkit.