Microsoft Unveils New Visual Studio Tools for AI Apps at Connect 2017

2017-11-17 10:13:14 admin 39

        Microsoft Nov. 15 took the wraps off new tools and cloud services that can help software developers seize the growing opportunity around artificial intelligence and the growing market of intelligent services it enables.

        Cloud computing, mobility, and the growing popularity of AI and internet of things (IoT) solutions are allowing developers to push the boundaries of what software can do using a richer palette of interconnected devices and types of data. During his opening remarks at the software giant's Connect 2017 conference in New York City, Scott Guthrie, executive vice president of Microsoft's Cloud and Enterprise division, told attendees that there has "never been a better time than today to be a developer."

        Helping developers capitalize on this opportunity, Microsoft made a rash of new announcements, including the preview release of Visual Studio Tools for AI. This extension for the Visual Studio IDE (integrated development environment) allows coders to build, debug and deploy deep learning and other AI software solutions      

        Visual Studio Tools supports multiple deep learning frameworks, including Caffe, TensorFlow and, of course, Microsoft's own Cognitive Toolkit. It extends Visual Studio's contextual editing capabilities and features an Azure Machine Learning integration that allows users to seamlessly transfer AI model training jobs to Microsoft's cloud if and when a little more processing power is required.

        The Visual Studio Tools for AI preview is available now. The extension supports Visual Studio 2017 and 2015 on 64-bit Windows operating systems (Windows 10 is recommended)

        Another preview release, Azure IoT Edge, allows developers to deploy and run intelligent services on edge devices. Backed by Azure Machine Learning, Azure Functions and Azure Stream Analytics, developers can run AI applications on the edge using containers and targeting devices with at least 128MB of memory. Relatedly, Microsoft has updated its Azure Machine Learning service to allow for the deployment and processing of AI models on such devices using Azure IoT Edge

        A Focus on Real-Time Developer Collaboration Microsoft is betting that creating a new generation of intelligent applications and services not only requires new AI-enabled tools but also new, more immediate ways of collaborating and sharing expertise.

        Accordingly, Microsoft unveiled its upcoming Live Share feature for both Visual Studio and Visual Studio Code, the company's lightweight, cross-platform code editor. Rather than submit an issue on GitHub, wait for a reply on Stack Overflow or use screen-sharing software, developers can now solve problems together using the new real-time, in-editor experience

        In an on-stage demonstration, Microsoft executives showed how Live Share creates a remote connection to a user's workspace using a web link that can be shared over Slack or other messaging platforms. One developer can use Visual Studio on Windows while another uses Visual Studio on a Mac and still arrive at a solution (some sort of voice chat solution helps the process along), each making changes that are reflected on one another's code editor in an experience that resembles the co-authoring features in Office

        Also on Nov. 15, Microsoft announced the general availability of Visual Studio App Center, a cloud-based application lifecycle solution for Windows, iOS, Android and macOS apps with support for Objective-C, Java, React Native, Swift and Xamarin. A preview version of Azure DevOps Projects, allowing organizations to quickly establish a DevOps pipeline, is also available

        Finally, Microsoft will soon let developers kick the tires on its new Connected Environment for Azure Container Service (AKS) offering. It allows Visual Studio and Visual Studio Code users to edit and debug applications running on Kubernetes, the market-leading open-source container orchestration technology