Jun 19, 2023

Microsoft Build 2023 – The Highlights

Microsoft Build Conference 2023 brought together developers, software engineers, IT professionals and businesses to launch the latest news on Microsoft's services and platforms. A notable focus was on Artificial Intelligence (AI) and the occasion proved to be a turning point for the businesses as it made several ground-breaking statements on the incorporation of AI into a variety of its products and services, ranging from the ground-breaking Windows 11 to the comprehensive Microsoft 365 suite and beyond.

Day 1 |Recap - Artificial Intelligence Takes the Center Stage

1. Revolutionizing Software Development: Microsoft's AI Copilots

In today's connected world, it's hard to imagine software that isn't linked to the internet. According to Kevin Scott, Microsoft's chief technology officer, the same will soon be true for software without an intelligent copilot assistant.

A copilot is an application powered by advanced AI and large language models that assists users with complex cognitive tasks. Whether it's writing a sales pitch, catching up on missed meetings, creating presentation images or planning a themed dinner party, a copilot can handle it all.

Nearly two years ago, Microsoft introduced the concept of a copilot with GitHub Copilot, an AI pair programmer that helps developers write code. Since then, copilot experiences have been integrated into various Microsoft products and services, such as Bing, Microsoft 365, GitHub, Dynamics 365, Microsoft Viva and Microsoft Security.

At the annual Microsoft Build developers conference, Microsoft announced the expansion of its copilot ecosystem. New additions include Copilot in Power BI, Copilot in Power Pages (in preview), Copilot in Microsoft Fabric(coming soon) and Windows Copilot (available for preview in June). To enhance the functionality of copilots, Microsoft introduced plugins, which enable copilots to interact with other software and services.

Microsoft's adoption of OpenAI's ChatGPT plugin standard allows interoperability across ChatGPT and Microsoft's copilot offerings. This means developers can now build plugins that work seamlessly across both business and consumer surfaces, including ChatGPT, Bing, Dynamics 365, Microsoft 365 and Windows Copilot. Bing will also serve as the default search experience for ChatGPT.

The plugin ecosystem is expanding rapidly. Bing now supports plugins for OpenTable, Wolfram Alpha, Expedia, Instacart, Kayak, Klarna, Redfin, TripAdvisor, Zillow and more. Developers can extend Microsoft 365 Copilot with plugins, including ChatGPT, Bing plugins, Teams message extensions and Power Platform connectors. Microsoft provides tools like theMicrosoft Teams Toolkit for Visual Studio Code and Visual Studio to simplify plugin development.

So how do plugins work?

They act as bridges between large language models and various data sources or systems. For example, a copilot might access a company's private data to answer questions about past legal issues. Plugins make this possible by connecting copilots with specific databases or back-end systems. Microsoft is releasing capabilities to facilitate the creation of plugins that work across its copilot surfaces. Developers can leverage tools like Visual Studio Code, GitHub Copilot, GitHub Codespaces and Azure AI to create, debug and deploy plugins.

Developers are at the forefront of copilot development. While Microsoft has already deployed generative AI technology across its products and services, most copilots will be built by developers who understand their users needs. The copilot development process starts with selecting a generative AI model like GPT-4 and providing a meta-prompt that defines the copilot's role and operation. Memory and plugin access are essential for context-awareness and data utilization. Orchestration capabilities allow copilots to execute complex tasks effectively.

Microsoft introduced new tools at Build to simplify copilot development. Azure AI Studio enables developers to ground conversational AI models on private data. Azure OpenAI Service provides organization-specific insights from data, text and images. Support for plugins in Azure Open AIService further extends the capabilities of large language models. Azure AI also supports Azure Machine Learning prompt flow, a tool for prompt engineering.

Responsible AI is a critical aspect of copilot development. Microsoft announced Azure AI Content Safety, a service that detects inappropriate content across images and text. It assigns severity scores to flagged content, enabling human moderators to take appropriate action.

2. Plugins for Microsoft 365 Copilot

The launch of Microsoft 365 Copilot earlier this year introduced generative AI models to assist developers in code writing. Now, Microsoft has introduced extensibility with plugins, allowing developers to integrate their applications and services into Microsoft 365 Copilot. These plugins leverage AI capabilities to interact with APIs from other software and services, enabling real-time information retrieval, business data incorporation and new calculations.

3. Microsoft Fabric - Unifying Data Analytics

In the era of digital transformation, data plays a crucial role. Microsoft Fabric, a unified analytics platform, combines all the necessary data and analytics tools businesses need. It integrates technologies such as Azure Data Factory, Azure Synapse Analytics and Power BI into a single comprehensive product. With Microsoft Fabric, organizations can harness the power of data and lay the foundation for the AI era.

4. AzureAI Studio - Empowering Next-Generation Applications

Azure AI Studio, a new AI tool set for Azure, aims to facilitate the building, running, and deployment of next-generation applications. This tool set simplifies the integration of external data sources into Azure Open AI and Azure Machine Learning, offering solutions like Semantic Kernel to analyse natural language text and generate semantic representations, capturing meaning.

Day 2 | Recap: Power BI, Azure Cosmos DB and More

The second day of Microsoft Build 2023 continued to bring exciting announcements, updates to their existing products and innovations. Let us dive into the highlights.

1. Azure Cosmos DB Enhancements - Optimized for Performance and Productivity

Azure Cosmos DB has established itself as the leading cloud database for generative AI and application development workloads. To further optimize cost, performance, and developer productivity, several enhancements have been introduced.

· Burst capacity: Leveraging idle throughput capacity from databases or containers, users can handle   traffic spikes more efficiently, ensuring smooth operations even   during high-demand periods.

· Hierarchical partitioning keys: With improved performance and more efficient strategies, Azure Cosmos   DB eliminates performance trade-offs by enabling the creation   and maintenance of secondary views of   data in containers. These views are used to serve queries that would otherwise be resource intensive.

 · Materialized views for NoSQL: Users can now create and maintain secondary views of their data in   containers, making it easier to serve queries that would otherwise    be too costly. This enhancement   opens new possibilities for efficient data retrieval and analysis.

2. Microsoft Azure Container Storage - Enhanced Performance and Reliability

Microsoft Azure Container Storage, a unique volume management service built specifically for containers, has introduced new features to enhance performance, reliability and workload experience.

· Volume snapshot: Users can now capture the point-in-time state of persistent volumes, allowing for   easy restoration or replication of data as needed.

· Increased scalability target for Persistent Volumes: This enhancement enables seamless scaling of   the storage footprint, ensuring that businesses can accommodate   growing data demands effortlessly.

3. Confidential Containers on Azure Kubernetes Service - Zero-Trust Deployments

Confidential Containers on Azure Kubernetes Service (AKS) provide a valuable solution for teams seeking to achieve zero-trust operator deployments. By running standard, unmodified containers in alignment with the open-source Kata Confidential Containers project, organizations can ensure secure operations within AKS.

These confidential containers seamlessly integrate with typical services used by applications running on AKS, including monitoring and logging. This integration ensures that sensitive data remains protected within a trusted execution environment.

4. Azure Deployment Environments - On-Demand Self-Deployment

Azure Deployment Environments are now generally available, empowering users to self-deploy the required environments on demand using infrastructure templates as code. With a dedicated portal, creating and managing environments becomes a streamlined process, providing a clear and simple interface for users. The portal also offers functionality to view, manage and activate Microsoft Dev Boxes, further enhancing the development experience.

5. GitHub Advanced Security for Azure DevOps - Native Automated Security Checks

To bolster code, secrets and supply chain security without disrupting workflows, GitHub Advanced Security is natively incorporating automated security checks into Azure DevOps. This integration provides the following protective measures. 

· Code scanning: Identifies vulnerabilities in source code and offers remediation guidance, ensuring the   development of secure applications.

· Secrets scanning: Prevents the introduction of secrets into code repositories, safeguarding sensitive   information.

· Dependency scanning: Discovers vulnerabilities within open-source dependencies and automate   update alerts.

6. Reimagining dev workstations with Microsoft Dev Box

Azure Deployment Environments, now generally available, allow developers to easily deploy required environments on demand using infrastructure-as-code templates. These enhancements include a dedicated portal with a user-friendly interface for creating and managing environments. The portal also provides tools to monitor, control, and provision dev boxes using Microsoft Dev Box. Additionally, development teams can choose between using Terraform or Azure Resource Management files to build their infrastructure-as-code templates. Terraform support is currently in preview and interested customer scan sign up for early access during the event.

· The developer workstation simplifies onboarding with self-service access to ready-to-use dev   environments. It offers pre-configured cloud-based dev boxes for instant   coding. They serve multiple   purposes like proof of concept, project isolation and handling concurrent tasks.

· Simplify team collaboration by managing ready-to-use dev box images. These images include tools for     various project types like desktop, mobile, gaming and IoT.   Allocate dev boxes based on Azure AD groups   with different access levels. Scale compute resources instantly for better performance.

· Efficiently manage and secure development workstations using Windows 365, Microsoft Intune   and Microsoft Endpoint Manager. Integrate these technologies for   unified management, enhanced   security and compliance. Control permissions and policies across dev boxes with Azure AD. Deploy   environments through the reliable   Azure network. Keep dev boxes updated with zero-day patches and   isolate compromised ones promptly.

Interested in hearing more?
Lets connect.