May 28th, 2025
0 reactions

Azure Cosmos DB TV – Everything New in Azure Cosmos DB from Microsoft Build 2025

Microsoft Build 2025 brought major innovations to Azure Cosmos DB, and in Episode 105 of Azure Cosmos DB TV, Principal Program Manager Mark Brown sits down with Kirill Gavrylyuk, Vice President and GM of Azure Cosmos DB, to break it all down.

Whether you’re developing AI-native apps, building with MongoDB, or just looking for better performance and observability, this episode covers all the latest updates—designed to help developers build faster, scale smarter, and integrate AI more seamlessly.

Let’s recap the key features and announcements covered in the episode and point you to the full Build 2025 sessions to dive even deeper.


🔍 Smarter Search and Indexing

Search just got a lot more powerful. Developers can now use Full-Text Search (GA), Hybrid Search, and Fuzzy Search to return better results—especially in AI applications.

Key highlights:

  • BM25 ranking support

  • Phrase and multilingual search

  • Integration with the .NET SDK’s LINQ provider and Entity Framework Core

  • Filtered DiskANN vector search (public preview), enabling scoped searches that combine full-text and vector techniques

▶️ Learn more in Tips for fast vector and full-text search with Azure Cosmos DB (DEM567)


🌐 Global Secondary Indexes & Resilience

Global Secondary Indexes are now in public preview, giving developers a way to query on any property—even if it wasn’t the partition key—across all regions. This improves developer flexibility without sacrificing performance.

Combined with Business-Critical Multi-region Write and Per Partition Failover, apps now get both resilience and ultra-low latency at global scale.

▶️ Learn more in Architecting highly resilient applications in Azure (BRK197)


🧠 Built for AI Agents

As Microsoft pivots toward the agentic web, Azure Cosmos DB is right at the center. New features integrate seamlessly with services like the Azure AI Agent Service and Azure AI Foundry.

Highlights include:

  • Bring-your-own thread storage

  • State management for AI agents

  • Fast reads/writes to support multi-agent systems

▶️ Explore these in
Building secure AI Agents with Azure Functions (BRK189)
Multi-agent apps with Semantic Kernel or LangChain & Azure Cosmos DB (LAB361)


🛠️ MongoDB Developers Get More

The vCore-based Azure Cosmos DB for MongoDB platform now offers:

  • New M10 and M20 tiers for cost-effective dev/test

  • Autoscale (GA) for dynamic throughput adjustment

  • Improved compatibility with MongoDB tooling and drivers

▶️ Learn more in Scale and secure MongoDB-compatible apps with Azure Cosmos DB (BRK202)


🚀 Fleet Management & Monitoring

For those managing multi-tenant apps, the new Fleet Management capabilities are game-changers. Group accounts into fleetspaces, share RU/s pools, and monitor performance across all tenants from a centralized view.

Complemented by:

  • Fleet analytics integration with Fabric

  • At-scale monitoring dashboards

  • Aggregated logs and improved diagnostics

▶️ Check it out in Design scalable data layers for multi-tenant apps (BRK212)


🧰 SDKs & Developer Experience

Developers now get a smoother experience across the board with:

  • New JavaScript SDK v4.0 (GA)

  • Better LINQ support

  • More observability built into SDK telemetry

These improvements make it easier to build intelligent, distributed applications—especially for full-stack and AI-focused devs.


🎮 Real-World Use Cases

The episode also covers how real-world organizations are already taking advantage of these innovations—from powering sports analytics with AI to defending the planet with agents in space (yes, really).

▶️ Don’t miss:
Beyond the Game: How AI and Data Are Powering the Future of Sports (BRK143)
Earth’s Defense with Hera: AI Agents Battle Planet Extinction Threats (BRK186)


🧭 Wrap-Up & Where to Watch More

Mark and Kirill wrap up the episode with a reminder: these features aren’t just “cool”—they’re built to solve real developer problems. Whether you’re scaling a global app, integrating AI, or reducing operational overhead, Azure Cosmos DB is delivering tools to help.

▶️ Watch the full episode: https://youtu.be/VIlnO3_jM3U
📺 Explore all Build 2025 sessions for Cosmos DB: https://aka.ms/AzureCosmosDB/Build2025

Leave a review

Tell us about your Azure Cosmos DB experience! Leave a review on PeerSpot and we’ll gift you $50. Get started here.

About Azure Cosmos DB

Azure Cosmos DB is a fully managed and serverless NoSQL and vector database for modern app development, including AI applications. With its SLA-backed speed and availability as well as instant dynamic scalability, it is ideal for real-time NoSQL and MongoDB applications that require high performance and distributed computing over massive volumes of NoSQL and vector data.

To stay in the loop on Azure Cosmos DB updates, follow us on XYouTube, and LinkedIn.

To quickly build your first database, watch our Get Started videos on YouTube and explore ways to dev/test free.

Author

Jay Gordon
Senior Program Manager

Jay Gordon is a Senior Program Manager with Azure Cosmos DB focused on reaching developer communities. Jay is located in Brooklyn, NY.

Mark Brown
Principal PM Manager

Mark is a Principal Program Manager on the Azure Cosmos DB team and is focused on making sure Azure Cosmos DB is the most developer friendly NoSQL database in the cloud. Mark is passionate about web development, cloud computing and growing the developer community around Azure and Cosmos DB.

Kirill Gavrylyuk
Vice President, GM, Azure Cosmos DB, Microsoft

Product Leader with experience of leading and growing multiple large scale cloud services from inception to maturity. Responsibilities include product strategy, business strategy, product execution, and leading engineering teams to deliver global cloud services. 20 years of experience building cloud services, distributed systems, frameworks & tools

0 comments