I work on the Azure Cosmos DB team. ❤️ Distributed systems and databases. Helping developers design for infinite scale.
Redmond, WA
Joined April 2008
- Tweets11.8K
- Following767
- Followers2.7K
- Likes8.9K
Hey @AlaskaAir not sure why but it's very difficult to access your site when I'm outside the US. Some countries in Asia in particular.
You may want to let your IT folks know you fly outside the US now. 🙂
I just realized I haven't opened Visual Studio in nearly two years. In fact, I haven't really done much C# in the last two years either. It's mostly been Python.
So, I uninstalled VS. Felt weird to do it.
And yes, I still work at Microsoft.
Hey @Uber_Support why is it so hard to reach your support folks on your app? I had a prebooked ride at airport. Driver picked up someone else leaving me to book another trip and pay a ton more for it. I need you to fix this. Not cool to pay more + cancel fees.
Hey @Uber why is it so hard to reach your support folks? I had a prebooked ride at airport. Driver picked up someone else leaving me to book another trip and pay a ton more for it. I need you to fix this. Not cool to pay more or cancel fees.
I realize this goes without repeating but Pete Hegseth sounds like an idiot every single time he speaks.
I would be embarrassed to ever show my face if I were him.
Mark Brown retweeted
Cosmos DB now runs natively in Microsoft Fabric! Blend NoSQL + SQL, stream real-time analytics into OneLake, and power BI with Copilot—all in one unified platform.
#CloudData meets #MicrosoftFabric 🧠
Learn more. 🔗 msft.it/6014s6J9I
ALT Screen capture of Microsoft Fabric dashboard. text reads: Announcing Cosmos DB in Microsoft Fabric (Preview) with New Capabilities! Learn more.
Mark Brown retweeted
🚀 Improve your #AzureCosmosDB query performance with Global Secondary Indexes!
GSIs let you run efficient queries across different partition keys—no manual sync, no heavy lifting.
🎥 Watch Mark Brown (@markjbrown) & Justine Cocchi explain: youtu.be/pl74ouvEe9Y
#NoSQL #MicrosoftBuild
Mark Brown retweeted
Looking for something new to learn? Python + Azure Cosmos DB might be your next move. With AI everywhere, data skills are more critical than ever. Your AI app can only be as good as your data.
Just released a 5-part series covering the fundamentals you actually need:
Data modeling: youtu.be/9sJGxigUido?feature…
Client configuration: youtu.be/D7GKPIx_ARg?feature…
Async operations: youtu.be/jmK5msNic9c?feature…
Batch operations: youtu.be/I0zH-Zb8d7M?feature…
Error handling & logging: youtu.be/e9K7_HPY2Aw?feature…
Everything's wrapped up in a template you can deploy immediately (or take apart for the bits you need): github.com/madebygps/cosmosd…
Mark Brown retweeted
Build C# multi-agent AI using Semantic Kernel + Azure Cosmos DB! 🤖
• Agent-based architecture
• Cosmos DB as vector & transactional store
• Real-world use cases + implementation insights
Watch → youtu.be/q0R5fPFdlrw #SemanticKernel #AzureCosmosDB #dotnet
Mark Brown retweeted
🚀 Everything new in #AzureCosmosDB from #MSBuild 2025!
Catch Kirill Gavrylyuk (@kirillg_msft) & Mark Brown (@markjbrown) as they break down all the latest:
🔹 AI agent integration
🔹 Global Secondary Indexes
🔹 Smarter search
🔹 MongoDB upgrades & more
📺 devblogs.microsoft.com/cosmo…
Mark Brown retweeted
Cost-Effective, Low Latency Vector Search
It looks like a pretty good read for developers looking to scale vector searches.
Proposal: Integrates DiskANN (a vector indexing library) inside of Azure Cosmos DB NoSQL (an operational dataset) that uses a single vector index per partition stored in existing index trees.
Benefit: "It supports < 20ms query latency over an index spanning 10 million of vectors, has stable recall over updates, and offers nearly 15× and 41× lower query cost compared to Zilliz and Pinecone serverless enterprise products."
It can further scale to billions of vectors with automatic partitioning.
If you're going to Microsoft #MSBuild come learn how to build multi-agent apps using Semantic Kernel in C# or #LangGraph in Python at our hands-on-lab.
build.microsoft.com/en-US/se…
There's another earlier delivery as well.
Mark Brown retweeted
Ep. 104 of #AzureCosmosDB TV is here!
🎥 High-Scale Data Processing over 1000s of Devices with Expanso. @markjbrown @aronchick
See how @ExpansoIO + Azure Cosmos DB handle 50M records in under a minute with 1,000 edge nodes 🌍
youtu.be/F2VMVjSxIHY
#EdgeComputing #DistributedSystems #NoSQL
Come check out the latest for @AzureCosmosDB and see the exiting things customers are building at Cosmos DB Conf. Streaming live NOW!!!
youtube.com/watch?v=qXSur9LI…
I just updated our Samples Gallery with our latest great sample apps.
1. Multi-agent sample using OpenAI Swarm and @AzureCosmosDB
2. Full-day workshop building multi-agent apps using LangGraph in Python or Semantic Kernel Agents in C#.
azurecosmosdb.github.io/gall…
We built a new 1-day workshop on building Multi-Agent AI Apps with @AzureCosmosDB
Learn using LangGraph (python) or Semantic Kernel Agents (C#), or both.
Code is here. Deploy/run full sample on main branch or use start branch for workshop.
github.com/AzureCosmosDB/ban…
Mark Brown retweeted
🚀 Scaling like counts, page views, API usage, or anything requiring high-volume increments? The Distributed Counter pattern keeps things accurate & fast in #AzureCosmosDB.
Learn how: youtu.be/InCb-fcqWWA?si=CKUV…
Starting my Monday with Outlook issues and a boatload of email to catch up on.
I have restarted Outlook 5 times and put over $40 in my swear jar in the last 30 minutes.
I can see right now, it's going to be a great week!