Microsoft Build 2023 conference has announced exciting new AI services and innovations for cloud developers. After being held entirely online for the last three years due to the pandemic, the conference will finally be held in Seattle in front of an audience in 2023. For those who cannot attend in-person, all Build lectures can be watched online via live stream or recording for free.
Azure AI Studio is a new web-based tool as part of the Azure OpenAI Service. It allows developers to create, customize, train, and deploy AI models. Developers can select from a model catalog that includes their own models and open-source models and combine various data sources graphically. Another web-based tool is the Prompt Flow in Azure AI Studio.
Azure Deployment Environments, released as a preview in October 2022, are now ready to use. This cloud service infrastructure is based on templates and enables development teams to establish an application. Microsoft is introducing a new framework called Semantic Kernel, which aims to make it easier to integrate Large Language Models (LLMs) into programming languages.
Microsoft has expanded its range of cloud-based Dev Box to include preconfigured Starter Developer Images. Dev Center is the new portal that helps manage Dev Boxes and Azure Deployment Environments. Microsoft also announced a simplified system called Unified Settings, facilitating the transfer of settings between different systems, including Dev Boxes.
Microsoft is taking the security scanning capabilities from GitHub to Azure DevOps by adding the feature of scanning for secrets and checking package dependencies for security vulnerabilities in Azure DevOps. It is now available in preview for all Azure DevOps users under the GitHub Advanced Security for Azure DevOps.
The Kubernetes-based service, Azure Container Apps, now has a new feature to start jobs for asynchronous tasks. The jobs can be time-controlled, event-based, started by a WebAPI call, or user action. Microsoft now offers two-year support for certain Kubernetes versions. They have also created their Linux distribution called Azure Linux optimized for use in the Microsoft Azure cloud platform.
Lastly, Microsoft has introduced a new section in Microsoft Learn dedicated to AI. Documentation on Azure OpenAI Services, the creation of AI-supported Power Apps and GitHub Copilot is available, among other things. Developers will also find links to Microsoft certifications and related training courses on AI. Moreover, the AI can support users in formulating a question with suggestions in the “Microsoft Q&A” forum area.