Voyage AI with MongoDB
In this skill badge, you’ll learn what vector embeddings are, how they’re created, and why they’re essential for powering modern AI applications and semantic search. You’ll explore how to store and query embeddings efficiently in MongoDB, and how to leverage MongoDB auto-embedding feature to generate and maintain embeddings in your database . Through hands-on labs, you’ll generate embeddings and apply Voyage AI models for both vector search and reranking, building a two-step search pipeline that boosts the relevance and context-awareness of your results. You’ll also examine key design choices—like model selection, chunking and context length, voyage-context-3, and embedding dimensionality—to balance search quality, latency, and cost for your workloads.
|
Upon completion of the Voyage AI with MongoDB skill and skill check, you will earn a Credly Badge that you are able to share with your network. |
Learning Objectives
Understand Vector Embeddings:
Develop a baseline understanding of embeddings and their related concepts in the context of application development.
Use Voyage AI Embedding Models and Rerankers
Use an appropriate Voyage AI embedding model and reranker.
Automate embeddings and rerank search results with MongoDB and Voyage AI
Leverage the MongoDB and Voyage AI integration to automate the embedding-generation process and implement reranking to improve search result relevance.
Who is this Course Good for?
If you are a developer, data scientist, or AI practitioner, the Getting Started with Voyage AI and MongoDB Skill Badge is designed for you. You may already be building search features using keyword-based approaches, but want more confidence that your system surfaces the most relevant content, even when user queries don't share exact words with your data. Whether you work on AI/ML teams, full-stack development teams, or data engineering teams, this skill will help you build intelligent search pipelines using MongoDB and Voyage AI's embedding and reranking models.
What to Expect in this Course
The skill begins with the foundational concepts behind vector embeddings: what they are, how embedding models generate them, and how they power semantic search. Unlike keyword search, which requires exact word matches, vector embeddings position similar concepts close together in a high-dimensional space, allowing a search system to return relevant results based on intent.
From there, you will store and search embeddings in MongoDB using two approaches: generating embeddings manually through the MongoDB Embedding and Reranking API, and using MongoDB's auto-embedding feature, which generates and indexes embeddings automatically as new data is ingested.
Next, you will explore reranking. While embedding models encode queries and documents separately for speed, reranking models evaluate them together for greater precision. You will learn when to add a reranker to your pipeline and apply Voyage AI's reranking model to a set of vector search results, observing how the most relevant document rises to the top.
The final lessons cover model selection and storage optimization. You will compare Voyage AI's model sizes from voyage-4-large to voyage-4-nano, explore chunking strategies and voyage-context-3 for precise retrieval over long documents, and use embedding dimensions and quantization to reduce storage costs at scale.
Throughout the skill, concepts are reinforced through detailed video lessons and a hands-on lab that grounds each topic in practical, real-world patterns. By the end, you will be equipped to choose the right Voyage AI embedding model for your use case, store and query vectors efficiently in MongoDB Atlas, and apply reranking models to build a semantic search pipeline that consistently surfaces the most relevant results.
Summary of the Course
- Describe what vector embeddings are and what they are used for.
- Understand why vector embeddings are important for AI applications and how they can be used to improve search.
- Describe how vector embeddings are created.
- Under Semantic Search, explain how vector embeddings are used to find similar concepts even when exact words aren't used.
- Learn how to store vector embeddings in databases and use them for efficient search.
- Generate vector embeddings using VoyageAI for your data.
- Understand different database options for vector search, including MongoDB.
- Learn how to leverage auto-embedding capabilities and distinguish between API-generated embeddings and database-managed embeddings.
- Understand what reranking models are and how they work.
- Distinguish between the use cases for vector search and reranking, and understand how they complement each other in a search pipeline.
- Create a two-step search process that combines vector search with reranking to improve the relevance of search results.
- Use Voyage AI's reranking models to enhance search results based on relevance and context.
- Identify the key factors — workload, data type, quality, latency, and cost — that influence which embedding model to choose.
- Define core concepts of chunking and context length.
- Describe operational challenges of maintaining context across chunks.
- Identify voyage-context-3 as a solution to chunking challenges.
- Understand how embedding dimensionality affects search quality, latency, and cost.
- Understand how to optimize dimensionality for your needs.
Emily Pope | Lead Curriculum Designer
Emily Pope is a Lead Curriculum Designer at MongoDB. She loves learning and loves making it easy for others to learn how and when to use deeply technical products. Recently, she's been creating AI and vector search content for MongoDB University. Before that, she's created learning experiences on databases, computer science, full stack development, and even clinical trial design and analysis. Emily holds an Ed.M. in International Education Policy from Harvard Graduate School of Education and began her career as an English teacher in Turkiye with the Fulbright program.
Manuel Fontan Garcia | Senior Technologist
Manuel is a Senior Technologist on the Curriculum team at MongoDB. Previously he was a Senior Technical Services Engineer in the Core team at MongoDB. In between Manuel worked as a database reliability engineer at Slack for a little over 2 years and then for Cognite until he re-joined MongoDB. With over 15 years experience in software development and distributed systems, he is naturally curious and holds a Telecommunications Engineering MSc from Vigo University (Spain) and a Free and Open Source Software MSc from Rey Juan Carlos University (Spain).
Parker Faucher | University Curriculum Engineer
Parker is a Curriculum Engineer on the Education team at MongoDB. Prior to joining MongoDB, he helped maintain a world class developer bootcamp that was offered in multiple universities. He is a self taught developer who loves being able to give back to the community that has helped him so much.
Aaron Becker | Technologist, Education
Aaron Becker is a Technical Trainer, Instructional Designer, and Training Manager who has worked in the tech sector for over 13 years. Before joining the Curriculum team at MongoDB, Aaron worked in DevOps at CircleCI, creating their first Certification course (CircleCI Associate Developer) and leading a team responsible for creating and managing the educational content for CircleCI Academy for external/customer training, as well as CircleCI University for internal team member training.
Prior to that, Aaron worked in data protection, redundancy, and security at Carbonite, where he headed up the Training team, created and delivered ILT training courses for Carbonite's Mid-Market and Enterprise level products, and assisted over 150 employees in earning Microsoft certifications.
Aaron enjoys writing, performing, recording, mixing and mastering music, playing video games, and writing biographical text in the third person.
In this skill badge, we're going to explore one of the most exciting areas in modern application development: Semantic Search powered by vector embeddings. Whether you're building a product recommendation engine, a document retrieval system, or a retrieval augmented generation application, this skill badge gives you the building blocks to get there. Traditional search relies on matching exact keywords. A user searches for something to wear in cold weather, but the most relevant product listing describes a thermal insulated jacket without using those exact words.
Keyword search would miss this entirely.
Vector embeddings bridge that gap. They turn data into numerical representations that capture meaning, so your search system can find relevant content even when the keywords don't match. Combine that with MongoDB's vector search capabilities and you have a foundation for a search pipeline that understands intent.
This skill badge is organized into seven lessons. We'll start with the foundation, what vector embeddings are and how embedding models generate them, and how to use them to perform semantic search. From there, we'll store and search embeddings in MongoDB.
We'll highlight a few ways to do this, including MongoDB's auto embedding feature, which handles embedding generation automatically as data is added. Next, we cover re ranking. You'll learn how reranking models differ from embedding models and when to add one to your pipeline.
Then, we'll put it into practice, applying Voyage AI's re ranker to a real set of results and watching the most relevant document rise to the top. The final three lessons focus on model selection and basic parameters.
We'll cover the quality versus latency trade off across Voyage AI's model sizes, how to document long documents using chunking, and how to manage storage at scale using embedding dimensions and quantization.
By the end of this skill badge, you'll understand how vector embeddings work, how to store and search them in MongoDB, and how to use Voyage AI's embedding models and rerankers to build a search pipeline that surfaces the most relevant results.
You'll learn through detailed videos and a hands on lab. When you're done, take a short skill check to demonstrate your knowledge. Pass it, and you'll earn an official Credly badge to share on LinkedIn. Let's get started.
