Tim Tully introduces a new course on building applications with vector databases through Pinecone. You'll build six apps: semantic search, RAG, recommender systems, hybrid search, anomaly detection, and facial similarity matching.
I'm delighted to introduce Building Applications with Vector Databases, built in partnership with Pineco and taught by Tim Turvey. In this course you learn how to use vector databases to build a wide range of applications, including searching for similar images, multimodal search involving simultaneously searching against text and images, anomaly detection, and recommender systems. At the heart of a vector database is the ability to create a collection of vectors and then query against that meaning, input a new vector and find similar ones. In the large language model world, the most common use of vector databases is RAG or Retrieval Augmented Generation, where you store embeddings of chunks of text in the vector database and use the database to retrieve chunks relevant to query. But as you learn from this course, there are many more applications that a vector database makes possible to build quickly. I'm delighted to introduce the instructor, Tim Tully, who's a board member at pineco. He's a seasoned technologist and had previously led teams of thousands of people as CTO of Splunk and as Vice President of Engineering at Yahoo. Tim was also the developer for the command line interface to Pinecone's Vector database and has seen firsthand a lot of what actually works and doesn't work for vector database use cases.
Thanks, Andrew. We have a packed agenda in this course. You build six apps powered by vector databases. We'll begin with a basic semantic search app using a Vector DB to find similar documents, then move on to a RAG application. Touch on recommender systems and something called Hybrid Search where you can work simultaneously with this dense and sparse vectors. Then apply what you've learned to anomaly detection in network logs and finally to a facial similarity app that can settle the biggest debate every parent has who does our kid look most like?
That's a lot of applications enabled by Vector dbs. After this course, I hope you have a lot of new ideas for what you can build yourself with Vector dbs. Please enjoy the course, which I'm confident will help you approach these applications from the right vector.
