Notes from Microsoft Build 2023

Introduction

Below are notes are taken from the MS Build 2023 2-day sessions that happened in Seattle this year that cover the sessions I have attended and my understanding and views on these topics. There are announcements made on major advancements happening in the Microsoft Technology stack that shift how we work and collaborate in a new and intuitive manner.

Keynote

The keynote from MSFT CEO Satya highlighted how the transformation of computers and the internet in the 70s and how they are now after 5 decades. He talked about the Mainframes to PCs to mobile today changing the way we work, day to day. The main emphasis of the keynote is on the co-pilot stack. Below is the screenshot that shows the Co-Pilot stack.

Co-Pilot

Windows Co-Pilot

There is a short video on how you can use your natural language to personalize your PC, such as

  • Changing the theme to dark
  • Installing / uninstalling programs
  • Preparing a logo and sending in teams
  • Much more

This is all embedded in Windows 11 OS.

Microsoft 365 Co-Pilot: This has been in action for the past few months. If you open an edge window, you must have already seen a widget that is personalized to the user, which gives an overview of what happened yesterday and frequent tasks.

Co-Pilot

Clicking on 'See more in Microsoft 365' takes a full summary of tasks run by your team. There is also a demo showing Co-Pilot creating a summary report from your emails and one drive files with the help of Syntex, also an auto-generated presentation with your email conversations from the teams/emails. With the co-pilot, the days are gone where the mundane tasks like taking meeting minutes and generating summaries will be taken care of by Co-Pilot. You must connect the dots, and Co-Pilot does the rest.

Power-Platform Co-Pilot

The idea of the Power Platform is to make development easier for everyone. With the introduction of Co-Pilot, it is becoming much easier. Co-Pilot not only gives you recommendations or solutions to your questions, but it also puts the sample design in your design, where you have the flexibility to make changes. You do not have to build your solution from the bottom down. There are cool demos shown in the Power-Platform co-pilot. Below are the references to dig more.

https://learn.microsoft.com/en-us/power-apps/maker/canvas-apps/add-ai-copilot

Below is the screen capture of how a new flow designer powered with the co-pilot feature.

Co-Pilot

Please note that the co-pilot is currently in preview and will be Generally Available in June.

Git-hub Code Spaces

Like Power-Platform co-pilot, the Git-hub Co-Pilot is already in action, where it generates a line of code, whether it is (C#, Java, YAML file, etc.) the moment you enter the comments it gives you code in intellisense format. Note that the Git-hub co-pilot comes with a $10 monthly subscription for individuals. More about the Git-hub co-pilot can be found here.

https://docs.github.com/en/billing/managing-billing-for-github-copilot/about-billing-for-github-copilot

Azure AI Studio

This is a full complete life cycle toolchain to build intelligent AI Apps. Many improvements have been made with inculcating Chat GPT in the Azure AI studio. This uses an Open AI standard RAG pattern (Retrieval Augmented Generation). It happens in 3 steps.

  • Retrieve the data from a data source
  • Inject / Augment the data in the model and train the model
  • Generate response

Azure AI Studio

There is also a short demo of how easy it is to create your company's own chat GPT application using a few clicks in Azure Studio. More about the AI studio enhancements can be found in the link below.

With these enhancements, the model is also monitored for the responses automatically for any harmful content. The AI standards' goal is to provide intelligent and robust applications while keeping safety and security as the principal rules.

https://azure.microsoft.com/en-us/blog/build-next-generation-ai-powered-applications-on-microsoft-azure/

Microsoft Fabric

Data is the fuel for innovations in AI. Microsoft Fabric is a one-stop shop to analyze data and provide real-time analytics. It is built on 3 modules.

  • Data Factory: To ingest the data from various data sources
  • Synapse: to store the data, analyze the data, observe the stream and pattern, and provide real-time analytics
  • Power BI: to create rich and visual dashboards

OneLake

There are many submodules integrated into this MSFT Fabric; More about Microsoft Fabric can be learned from the link below.

https://learn.microsoft.com/en-us/fabric

The Era of Co-Pilot

https://build.microsoft.com/en-US/sessions/bb8f9d99-0c47-404f-8212-a85fffd3a59d?source=sessions

More on M365 Co-Pilot and how you can customize and extend the existing functionality of Co-Pilot to meet the needs using custom plugins. The data that existing various sources from your Tenant (Teams, SharePoint, OneDrive, Sway, Dataverse, etc.) are

  • Discovered
  • Synthesized
  • Summarized

And you can target this data to various sources (link Bing for business, Windows widgets) by using custom plugins. These custom plugins act on the user's behalf, just like delegated permissions in Azure which means the custom plugins act on behalf of the user in the user's context. You can also use the same plugins to use in Power Platform applications. Also, the plugins support industry standard Open Auth 2.0 Authentication.

Azure Dev-boxes

These are in the preview state. It gives users provides a preconfigured machine that is in a ready-to-code state. It comes pre-packaged with all the required software and network setup.

https://azure.microsoft.com/en-ca/products/dev-box/

Azure Deployment Environment

The deployment environments help to spin up the infrastructure to test the applications in real-time environments. It also provides templates to set up a dev-test and integration with code repositories to set up CI/CD pipelines.

They also showed a short video of how our parent company GM took advantage of the Azure Dev-Box and Azure Deployment Environments to accelerate the development of edge solutions and shipping to end users.

https://azure.microsoft.com/en-us/products/deployment-environments/

azd CLI

I initially thought that this was Azure CLI, but I was wrong. This is azd cli stands for (Azure Developer CLI), which helps to directly ship the code from your local environment to Azure with friendly commands and best practices. It involves the following steps.

  • Clone the code
  • Add your customization
  • Set up CI/CD
  • Deploy to Azure

More about the azd cli can be found below.

https://learn.microsoft.com/en-us/azure/developer/azure-developer-cli/overview

References

Build Opening: https://build.microsoft.com/en-US/sessions/49e81029-20f0-485b-b641-73b7f9622656?source=sessions

Teams Toolkit

More improvements have been done on Teams Toolkit V5 for Visual Studio Code, where the following steps.

  • Build
  • Debug
  • Deploy

It is made much simpler and supports side-loading of plugins. Some of the interesting points

  • Debugging in real time is much simpler now by configuring the manifest file.
  • The authentication code which we used to write has now been given as a template, and you just must update the variables for Tenant ID and a few more.
  • Teams Toolkit has built-in ACE (Adaptive Card Extension), which you can readily use in your apps.

Semantic Index for Co-Pilot

The intelligent feature, based on MS Graph, runs in the background across the various sources in the M365 tenant. It analyses the data from individuals and tries to build relationships, and helps to power the M365 co-pilot to bring improvements and ease to your daily tasks and change the way we work. This index is like an artificial Brian that brings intelligence to your work. More about the semantic index can be learned here.

https://techcommunity.microsoft.com/t5/microsoft-sharepoint-blog/microsoft-search-connecting-you-to-knowledge-and-expertise/ba-p/3821892

A short demo has been shown using VS Code and Teams Toolkit how a message extension powered with Semantic Index for Co-Pilot is being developed in less than 10 minutes. There is a session on the Teams tab building from the below session, which also covers the latest features.

Learn Live: Build a bot and Teams tab app with Teams Toolkit for Visual Studio Code (microsoft.com)

Microsoft Syntex

AI-powered content processing tool that understands your data and classifies it and processes the data, and extracts the key information. There is a short demo on how a contract agreement has been generated from the existing sales sheet and emails within a matter of minutes, which normally takes a couple of hours to generate. More about the syntax can be learned from the learning module.

https://learn.microsoft.com/en-us/microsoft-365/syntex/syntex-overview

Microsoft Mesh

Currently, teams meeting is in 2D, and most of us, after some time, could feel video fatigue. MSFT had made an announcement in build related to the interactive and immersive meetings using MSFT mesh, where you can customize the meeting theme, and you can end up meeting in person with your personal emoji being present in your own space using VR devices and Teams client on your PC. More information about MSFT mesh can be found below.

Azure AI Studio

https://www.microsoft.com/en-us/mesh#feelpresence

Dev Home for Windows

Dev home is a new control center used to set up the Dev machine with the desired state with a record time. Imagine you want to onboard a new employee or set up a developer machine for your team, etc., and all these tedious tasks can be handled in Dev Home with a few clicks and in a matter of minutes. All the setups can be finished, and the machine is ready with the pre-configuration setup of the required modules and packages. More about the Dev Home can be found below.

https://learn.microsoft.com/en-us/windows/dev-home/

There is also a demo on how to set up the dev box in a few clicks.

Set up your dev machine in record time with WinGet and Desired State Configuration (microsoft.com)

Project Silica

Storing massive amounts of data is expensive. The amount of data produced grows much more rapidly than capacity. The data is stored in glass instead of magnetic discs. For more information on Project Silica, you can refer to the complete session on 'Inside Azure Innovations" from the references section.

Project Copacetic

If there are some critical security patches and vulnerabilities that need to be applied at the server / VM (Virtual Machine) level, it can be achieved by several automation jobs, such as Run books, Desired State Configurations, PowerShell scripts, etc. Imagine if these security patches and vulnerabilities must be addressed in the container images; it is certainly difficult. In theory, an image is built on top of the OS; if the OS has security issues, then the image will have it. To address these difficult tasks, Project Copacetic was introduced by the Open-Source community. The tool known as 'copa' is a CLI tool written in Go, and it can be used to patch the container images without having 'Downtime' to run applications and avoid rebuilding the fresh images from OS. More about Project Copacetic can be found here, and you can check the live demos from the references section 'Inside Azure Innovations' presentation.

Some Interesting Sessions

Below are some of the sessions that caught my attention.

There are many more sessions available. Please login to https://aka.ms/build for more information.

Conclusion

There is an interesting closing note from MSFT CEO Satya, why do we BUILD? We like to increase the level of education and prosperity for every human being; that is why we BUILD, that is why we innovate, and that is why technology exists. And there is no better time than now to become a developer, even though you have not written a single line of code. That is right, you heard it right. The Co-Pilot and other AI modules will help in each step, right from designing the app to deployment and finally shipping it to end users. What are you waiting for?


Similar Articles