Azure Cosmos DB Conf 2026 streamed live on April 28, 2026 from 9:00 AM to 2:00 PM PDT. All times shown in Pacific Daylight Time. Every session is now available on demand.
Kirill Gavrylyuk, Guillermo Rauch, Jonathan Lee, Andrew Liu, Steve Berg
Kirill hosts a keynote conversation with leaders from Vercel, OpenAI, Walmart, and AMD on building modern applications, AI, and data at planetary scale.
Mick Feller, Distinguished Software Engineer at The ODP Group (Office Depot), on powering an enterprise AI Personal Assistant with Azure Cosmos DB.
Sid Anand, Technical Fellow at Walmart, uses the Jetsons' Spacely Sprockets as a fun lens for the best practices behind always-on, low-latency, cost-efficient Cosmos DB apps.
Three production-ready memory patterns in Azure Cosmos DB that help AI systems retain context without blowing up token usage.
A deep dive into Change Feed internals, consumption models, lease management, and how to debug and scale it in production.
A sneak preview of a portable Java SDK that runs unchanged across Azure Cosmos DB, Amazon DynamoDB, and Google Cloud Spanner.
Andrew Ruffin, Senior Product Manager at AMD, on the modern AMD EPYC processors that power Azure Cosmos DB across the globe.
How to unify semantic caching, Change Feed coordination, and optimistic concurrency into a single Agent Memory Fabric that slashes multi-agent costs.
Build a retail store database once with MongoDB-compatible APIs and deploy it unchanged on-prem via Kubernetes and fully managed on Azure.
Johnny Halife, Partner at SOUTHWORKS, on using Azure Cosmos DB as the operational backbone for cloud-native, event-driven customer builds.
A complete tour of the Azure Cosmos DB query engine and indexing system — routing, index types, query metrics, and cost-saving optimizations.
Jess Ramos, founder of Big Data Energy Analytics and LinkedIn Top Voice in Data & AI, on building a community for 500,000+ data and tech professionals.
A whiteboard walkthrough of Azure Cosmos DB's internals — replica sets, consistent hashing, Request Units, and RU pooling across a global fleet.
Strategies for partition keys, schema design, storage estimation, and managing cross-partition queries before the choices become painful to fix.
Guillermo Rauch, CEO of Vercel and creator of Next.js, joins Kirill Gavrylyuk on building a developer cloud for the AI agent era.
Jonathan Lee from OpenAI's online storage team joins Kirill Gavrylyuk on what it takes for a database to keep up with OpenAI.
Steve Berg, CVP & GM Server CPU Cloud at AMD, joins Kirill Gavrylyuk on the AMD EPYC silicon powering Azure Cosmos DB across the globe.
Use Cosmos DB as a coordination layer for locks, leases, leader election, and saga orchestration — and understand the consistency tradeoffs at scale.
A real-world walkthrough of diagnosing hot partitions, over-indexing, and throughput issues — and the optimizations that restored stable performance.
Build a Python MCP server that authenticates users via Entra ID and stores per-user data securely in Azure Cosmos DB.
A cohesive framework for identity-first access, isolation, encryption, and data protection in cloud data workloads — built from real customer scenarios.
Combine event sourcing, Change Feed, and event-driven integration to build an auditable real-time fraud detection system across heterogeneous data sources.
Extend a working AI agent with agentic RAG using Cosmos DB hybrid search — vector plus full-text — so it retrieves what actually matters.
A tour of migration options and tradeoffs for moving MongoDB workloads to Azure DocumentDB.
Repeatable data-modeling, TTL, Change Feed, and observability patterns that cut Cosmos DB costs without sacrificing performance or velocity.
A complete Cosmos DB dev workflow — emulator, testing, CI/CD — supercharged with context files and prompting techniques for AI coding assistants.
A prompt-by-prompt playbook for migrating AdventureWorks and an ASP.NET EF Core app to .NET 9 on Cosmos DB using GitHub Copilot and Agent Kit.
Production patterns for building loosely coupled, event-driven microservices with Cosmos DB Change Feed as the domain event backbone.
Build chat history and semantic cache containers that give AI agents persistent, context-aware memory while cutting tokens and latency.
A decision framework for choosing between the Data Migration Tool, Azure Data Factory, and the Spark connector — plus capacity planning for bulk loads.
