Q&A

Build Next-Gen AI Apps with .NET and Azure

• By David Ramel
• 10/16/2024

.NET developers are in a great place these days when it comes to AI, thanks to Microsoft's leadership position stemming from a massive investment in generative AI pioneer OpenAI.

With that lead in AI, Microsoft has kept its dev tooling abreast, along with supporting platforms like .NET and the Azure cloud.

So there couldn't be a better time to get started with coding modern, cloud-based AI-infused applications in the Microsoft camp.

But how to get started?

Well, learning directly from a Microsoft AI expert would be a good place to start.

"We'll show you how to build cutting-edge, next-generation AI applications using .NET and Azure," said Jordan Matthiesen in describing his upcoming developer conference presentation aptly titled "Build Next-Gen AI Apps with .NET and Azure. "

"We'll guide you through the end-to-end process, from development to production, revealing the powerful tools and frameworks available for building intelligent applications."

The senior product manager at Microsoft will be sharing his expertise as part of the Artificial Intelligence Live! developer conference coming to Orlando next month, just one part of the multi-conference Live! 360 umbrella event for developers and IT pros, with conferences for Visual Studio, Cloud & Containers, Cybersecurity & Ransomware and more.

"Learn how to harness the power of the cloud to scale your AI workloads and discover how you can bring these AI workloads to your existing applications," Matthiesen said, promising that attendees will learn:

• What you can do with .NET and AI today
• Hands-on tips for how you can incorporate AI into your existing .NET apps
• About the .NET roadmap for working with AI in .NET 8, .NET Aspire, .NET 9 and beyond

Want to learn more about the session? Here's a short Q&A:

PureAI: What inspired you to focus on combining AI with .NET and Azure for this presentation?
Matthiesen: I work on the .NET team at Microsoft, focusing on how we can make AI development a great experience for .NET developers.

I've spoken with hundreds of developers and know that .NET devs are interested and excited about opportunities to use AI in their apps, but maybe aren't sure quite how to get started.

What are some key benefits of using .NET for AI app development compared to other frameworks?
.NET developers can use the languages, tools, and design patterns they already know, to build solutions using AI and large language models (or small local models). This makes it easier to enhance existing apps with new AI-driven features, as well as explore new product ideas without having to also learn a new language and ecosystem. I also find that the rich tooling out there for .NET, such as code analysis functionality provided by Roslyn, really helps me develop solutions faster with fewer errors and fewer rounds of build/debug/edit cycles.

Could you highlight some specific tools and frameworks within .NET and Azure that are particularly powerful for AI applications?
Semantic Kernel is a great library for .NET developers, providing a core set of APIs for working with generative AI development that also feels familiar to .NET devs. It also provides many "plugins," or connectors, for working with different AI models and data solutions.

Both the OpenAI SDK and the Azure OpenAI SDK are also invaluable for working directly with OpenAI models -- hosted either publicly or on Azure alongside your existing infrastructure.

We also just released a preview of Microsoft.Extensions.AI, a set of core .NET libraries that provide C# abstractions for working with AI services.

How does Azure facilitate the scaling of AI workloads in .NET applications?
Azure AI Services enable developers to build AI solutions using models from OpenAI, Microsoft, Meta, and many others, including thousands of models hosted on Hugging Face. These services also support APIs to work with Search, Speech, Vision, document processing/entity extraction, and monitoring for content safety.

These services, along with existing services like Azure App Services, Azure SQL Server, Azure Functions and so on can be used by any .NET workload and enable cloud-level scaling, performance, and security.

What's the very first thing you do if you're a dev in a .NET shop and you're asked to update the company's cloud wares with cutting-edge AI?
Jump right in and start "tinkering" -- experiment with the technology. I like to start with an existing sample application to get an idea of what's possible. Look for videos & blog posts online that talk about the fundamental concepts, as well as more advanced scenarios. Then start modifying the sample as you learn more, integrate some of your existing data to see what happens. Eventually, this can turn into a prototype for your application, but first you want to focus on understanding what this new tech can do.

Can you tease one sneak peek into the .NET roadmap you will share with attendees about what to expect in terms of AI features in .NET 9 and beyond?
We've got some good stuff in the works to make it easier to get started building .NET-based AI solutions on your local device -- including making this easier to do with .NET Aspire. I'll be showing off some new "bleeding-edge" work from our team that will be newly announced near the time of my session.

What resources would you recommend for those looking to delve deeper into this topic and prepare for your session?
The .NET + AI documentation is a great place to go to learn some of the core concepts and find code samples. There's also a detailed sample application eShopSupport you can use to explore use cases for AI that include chat, sentiment analysis, content generation, and more; as well as development workflows for local development and testing/evaluation. Beyond these official resources, I maintain an unofficial curated list of .NET + AI resources.

Note: Those wishing to attend the conference can save hundreds of dollars by registering early, according to the event's pricing page. "Save $300 when you register by the Early Bird deadline of Oct. 25!" said the organizer of the event, which is presented by the parent company of Pure AI.