
James Codella, PhD
Principal Product Manager -
Microsoft, Azure Cosmos DBJames is a Principal Product Manager working to enable planetary scale applications with Azure Cosmos DB. Specifically, he focuses on the Cosmos DB Query Engine and leads AI product work including state-of-the-art vector indexing and search with DiskANN, text and hybrid search, and other advanced information retrieval methods.
In a previous life, James was a machine learning researcher. He holds a B.S. and an M. Eng. degree from Cornell University, and a Ph.D. in Operations Research from The University of Wisconsin – Madison. He currently lives near New York City with his wife, Angela, and their dog, Statler.
Sessions Presenting
Fuel AI innovation with Azure Databases
On Demand
In Chicago + Online
Data is the fuel for innovation, driving the development of transformative applications and artificial intelligence (AI). This session will explore how harnessing the power of data across the Azure databases portfolio can lead to groundbreaking advancements that enhance operational efficiency, deliver personalized user experiences, and revolutionize the way we interact with technology.
Serverless GenAI: building scalable RAG apps with Azure Cosmos DB
On Demand
In Chicago + Online
In this session, we’ll explore advanced strategies for scalable serverless RAG applications using Azure Cosmos DB and DiskANN. Topics include architecture simplification, data synchronization, scalability challenges, and multi-tenancy. We’ll discuss the latest developments in vector search, full-text search, and hybrid search in Azure Cosmos DB, showcasing real-world use cases from our customers like EY. Learn how to get started easily with LangChain and Semantic Kernel.