From Data to Intelligent Agents: Building Trusted Enterprise AI with Microsoft AI Foundry with Shubhangi Goyal [MVP]

Power Pages - Simply Explained

Welcome to another episode of Knowledge Nuggets with Mirko Peters. In this episode, we're exploring Microsoft Power Pages, Microsoft's low-code platform for building secure, external-facing business websites. Modern organizations need more than marketing websites. They need customer portals, partner hubs, supplier dashboards, volunteer registration sites, and self-service platforms that connect directly to business data. Traditionally, building these solutions required web developers, expensive hosting, and months of development. Power Pages changes that by allowing organizations to build secure, data-driven websites that integrate seamlessly with Microsoft 365 and Dataverse—without starting from scratch. WHAT IS POWER PAGES? Power Pages is Microsoft's low-code platform for building external business websites. Unlike Power Apps, which focuses on internal employee applications, Power Pages is designed for users outside your organization, including: * Customers * Partners * Suppliers * Vendors * Contractors * Citizens * Volunteers Think of Power Pages as the secure front door to your business data. Visitors can securely log in, submit forms, check information, update records, and interact with your business processes while Microsoft manages the hosting, infrastructure, security, updates, and scalability behind the scenes. Rather than creating static websites, Power Pages creates interactive business portals connected directly to live enterprise data. FROM POWER APPS PORTALS TO POWER PAGES Power Pages wasn't built from scratch. Originally known as Power Apps Portals, Microsoft rebranded the product in 2022 to give it its own identity within the Power Platform. Today, Power Pages stands alongside: * Power Apps * Power Automate * Power BI * Microsoft Copilot The redesign introduced: * A modern Design Studio * Simplified licensing * Improved security * Better development experience * AI-assisted website creation Existing Power Apps Portal customers were automatically upgraded without rebuilding their sites. DATAVERSE: THE FOUNDATION Every Power Pages website is built around Microsoft Dataverse. Dataverse stores business information such as: * Customer accounts * Orders * Support tickets * Inventory * Projects * Service requests Power Pages acts as the presentation layer while Dataverse stores the underlying business data. When a customer submits a support request, updates their profile, or checks an order status, they're interacting directly with Dataverse through a secure server-to-server connection. Because Power Pages connects directly to live business data, organizations eliminate manual imports, spreadsheets, and duplicate data entry. Microsoft has also introduced Native Dataverse Authorization, allowing external users to receive security permissions directly within Dataverse, providing stronger security, improved auditing, and more granular access control. BUILDING SITES WITH DESIGN STUDIO Power Pages includes a visual Design Studio that simplifies website creation. The experience is divided into four primary workspaces. Pages Create pages visually using drag-and-drop controls, including: * Text * Images * Forms * Lists * Navigation * Sections Every form can connect directly to Dataverse tables. Styling Customize: * Branding * Fonts * Colors * Themes * Layout Most websites require little or no CSS customization. Data Manage: * Dataverse tables * Forms * Views * Relationships Without leaving the Design Studio. Setup Configure: * Site settings * Authentication * Integrations * AI capabilities * Go-live configuration Everything needed before publishing your website. SECURITY BY DESIGN Security is one of Power Pages' strongest features. New sites are private by default, preventing accidental exposure before administrators intentionally publish them. Microsoft provides enterprise-grade protection including: * Web Application Firewall (WAF) * Content Delivery Network (CDN) * DDoS protection * Microsoft Entra ID integration * SSL certificates * Managed hosting Authentication supports providers including: * Microsoft accounts * Google * LinkedIn * OpenID Connect * SAML identity providers Organizations define Web Roles that determine which users can access pages and which records they can view or modify. Combined with Table Permissions, Power Pages enforces row-level security so users only see the business information they are authorized to access. Microsoft also introduced Deep Scan, which analyzes websites for security weaknesses and recommends improvements before production deployment. LICENSING Power Pages uses standalone licensing separate from Power Apps. Pricing is based primarily on: * Authenticated users * Anonymous website visitors Organizations can start with a free 30-day trial that includes Dataverse storage and full development capabilities. Because Microsoft manages hosting, infrastructure, maintenance, security, and updates, organizations avoid the operational complexity of traditional web hosting platforms while benefiting from predictable cloud pricing. Become a supporter of this podcast: https://www.spreaker.com/podcast/m365-fm-modern-work-security-and-productivity-with-microsoft-365--6704921/support [https://www.spreaker.com/podcast/m365-fm-modern-work-security-and-productivity-with-microsoft-365--6704921/support?utm_source=rss&utm_medium=rss&utm_campaign=rss].

Power Apps - Simply Explained

Welcome to another episode of Knowledge Nuggets with Mirko Peters. In this episode, we're exploring Microsoft Power Apps, Microsoft's low-code platform for building custom business applications without traditional software development. Every organization has processes that deserve a dedicated app—equipment tracking, expense reporting, employee onboarding, inspections, approvals, or customer management. Traditionally, building these applications required software developers, long development cycles, and significant budgets. Power Apps changes that by allowing business users, citizen developers, and IT professionals to build secure business applications using visual tools, familiar Excel-style formulas, and deep integration with Microsoft 365. WHAT IS POWER APPS? Power Apps is part of the Microsoft Power Platform, alongside: * Power Automate * Power BI * Power Pages * Microsoft Copilot Its purpose is simple: build business applications quickly without writing thousands of lines of code. Developers design screens visually by dragging controls such as buttons, galleries, forms, text boxes, and images onto a canvas. Logic is created using Power Fx, a formula language inspired by Microsoft Excel, making it approachable for users who already work with spreadsheets. Power Apps connects directly to Microsoft 365 services including SharePoint, Teams, Outlook, OneDrive, and Dataverse, while also supporting hundreds of external systems through connectors. LOW-CODE DOESN'T MEAN NO-CODE Power Apps is often described as a low-code platform. Rather than writing complex C# or JavaScript applications, users build solutions visually while adding simple formulas only where needed. If you've used Excel formulas such as IF(), SUM(), or FILTER(), you'll quickly recognize the Power Fx syntax. The goal isn't to eliminate coding completely—it is to dramatically reduce the amount of technical knowledge required to create professional business applications. This allows departments to solve their own business problems instead of waiting months for traditional software development projects.  CANVAS APPS VS. MODEL-DRIVEN APPS Power Apps offers two primary application models. Canvas Apps Canvas Apps begin with a blank screen. Developers decide exactly where every button, image, gallery, and control appears, providing complete flexibility over the user experience. Canvas Apps work best for: * Mobile applications * Inspection forms * Approval apps * Dashboards * Simple business processes * Task-specific applications They're ideal when the user experience matters most. Model-Driven Apps Model-Driven Apps start with the data rather than the interface. Developers define tables, relationships, and business rules, while Power Apps automatically generates forms, navigation, and views. Model-Driven Apps are perfect for: * CRM systems * HR portals * Asset management * Project tracking * Customer databases * Complex enterprise applications Instead of designing every screen manually, developers focus on modeling the business data. DATAVERSE: THE FOUNDATION Behind many Power Apps solutions is Microsoft Dataverse. Dataverse is Microsoft's cloud-native business database built specifically for the Power Platform. It provides: * Secure tables * Relationships * Business rules * Role-based security * Data validation * Enterprise scalability Canvas Apps can connect to Dataverse or hundreds of other data sources. Model-Driven Apps require Dataverse as their primary data platform because they depend on its structured business data model. As organizations grow, Dataverse often becomes the preferred data source because it provides significantly more governance and scalability than spreadsheets or SharePoint lists. CONNECTING TO BUSINESS DATA One of Power Apps' greatest strengths is connectivity. Through more than 300 connectors, applications can integrate with: * SharePoint * Microsoft Teams * Outlook * OneDrive * Excel * SQL Server * Salesforce * SAP * Oracle * Google services * Dropbox * REST APIs * Custom business systems Instead of replacing existing applications, Power Apps extends them by creating modern business interfaces that connect directly to current data sources. MICROSOFT 365 INTEGRATION Power Apps fits naturally into the Microsoft ecosystem. Applications can run: * Inside Microsoft Teams * In any web browser * On iOS and Android devices * Embedded within SharePoint pages Security is managed through Microsoft Entra ID, meaning users authenticate using their existing Microsoft 365 accounts. Administrators control permissions using familiar Microsoft identity management tools while the Power Platform Admin Center provides centralized governance, environment management, and Data Loss Prevention policies. This deep integration allows organizations to deploy business applications without introducing separate identity systems or security models. BUILDING YOUR FIRST APP Creating a Power App is remarkably simple. Users can begin with an Excel spreadsheet or SharePoint list. Power Apps automatically generates: * A Browse screen * A Details screen * An Edit screen From there, developers customize colors, branding, layouts, business rules, and navigation while Power Apps handles much of the underlying application framework automatically. What once required weeks of custom software development can often be completed in less than an hour. AI AND COPILOT IN POWER APPS  Power Apps has changed dramatically with the introduction of Microsoft Copilot. Instead of manually designing every application, developers can now describe the app they want using natural language. For example: "Create an expense approval application with manager approvals and monthly reporting." Copilot can generate: * Screens * Navigation * Tables * Business logic * Forms * Power Fx formulas Copilot also assists users inside running applications by summarizing records, generating visualizations, answering questions about business data, and helping users find information faster. With the 2026 Wave 1 release, Copilot became generally available for Model-Driven Apps and introduced autonomous AI agents capable of coordinating routine business activities with minimal user intervention. Become a supporter of this podcast: https://www.spreaker.com/podcast/m365-fm-modern-work-security-and-productivity-with-microsoft-365--6704921/support [https://www.spreaker.com/podcast/m365-fm-modern-work-security-and-productivity-with-microsoft-365--6704921/support?utm_source=rss&utm_medium=rss&utm_campaign=rss].

Power Automate - Simply Explained

Welcome to another episode of Knowledge Nuggets with Mirko Peters. In this episode, we're exploring Microsoft Power Automate, one of the most powerful productivity tools in the Microsoft ecosystem. Many people think Power Automate is simply a workflow tool or "that flow app" hidden somewhere inside Microsoft 365. In reality, it's a business automation platform that connects hundreds of applications and services, eliminating repetitive manual work without requiring you to write code. Whether you're automating approvals, processing forms, managing documents, or integrating business systems, Power Automate helps you work smarter while saving countless hours every week. WHY POWER AUTOMATE EXISTS Every organization spends countless hours performing repetitive tasks. Employees manually: * Copy information between systems * Forward emails * Move documents * Update spreadsheets * Create tasks * Request approvals * Send notifications While these activities are essential, they rarely create business value. They consume time, introduce human error, and prevent employees from focusing on more meaningful work. Before Power Automate, automating these processes often required expensive enterprise software or custom development. Microsoft changed this by creating a platform where anyone can automate business processes using a visual interface instead of writing code. THE CORE CONCEPT: TRIGGERS AND ACTIONS At its heart, Power Automate is built around two simple concepts. A Trigger determines when automation begins. Examples include: * A new email arrives * A SharePoint file is created * A Microsoft Form is submitted * A Teams message is posted * A scheduled time occurs Once the trigger fires, Power Automate executes one or more Actions. Actions can include: * Sending emails * Creating files * Updating SharePoint lists * Posting Teams messages * Creating Planner tasks * Updating Excel rows * Calling external APIs Everything inside Power Automate is built from these simple building blocks. The entire platform can be summarized as: When this happens... do that. THE THREE TYPES OF FLOWS Power Automate supports three primary flow types. Automated Flows Automated flows run whenever an event occurs. Examples include receiving a customer email, uploading a document, or creating a new SharePoint item. Once configured, these flows run entirely in the background without user interaction. Instant Flows Instant flows begin when a user manually starts them. These can be launched from: * Microsoft Teams * SharePoint * Mobile devices * The Power Automate app They're perfect for actions such as requesting approvals or generating reports on demand. Scheduled Flows Scheduled flows execute at predefined intervals. Examples include: * Daily reports * Weekly reminders * Monthly backups * Recurring maintenance * Data synchronization These automations operate on timers rather than user activity or business events. CONNECTORS: THE REAL SUPERPOWER Power Automate becomes incredibly powerful because of Connectors. A connector allows Power Automate to communicate with another application or service. Microsoft currently provides more than 1,400 connectors, including: * SharePoint * Outlook * Teams * OneDrive * Excel * Dynamics 365 * Salesforce * Adobe * DocuSign * Google Workspace * Mailchimp * Twitter * Thousands of additional services Each connector exposes its own triggers and actions, allowing organizations to connect virtually every part of their technology landscape. If no connector exists, developers can even create custom connectors for their own APIs. CLOUD FLOWS VS. DESKTOP FLOWS Power Automate actually includes two different automation technologies. Cloud Flows Cloud Flows run entirely inside Microsoft's cloud infrastructure. They automate cloud services using APIs and connectors without requiring your computer to remain online. These are ideal for Microsoft 365, Dynamics 365, SharePoint, Teams, Outlook, and modern SaaS platforms. Desktop Flows Desktop Flows use Robotic Process Automation (RPA). Instead of calling APIs, they literally interact with software like a human user by: * Moving the mouse * Clicking buttons * Typing text * Reading screen content * Copying information * Opening applications Desktop flows are especially valuable for older legacy applications that don't provide APIs or modern integration capabilities. BUILDING COMPLETE BUSINESS PROCESSES The true value of Power Automate comes from combining multiple actions into complete business workflows. Imagine a customer submits a contact form. Power Automate can automatically: * Save the submission to SharePoint * Find the correct salesperson * Send a Teams notification * Create a Planner task * Log the interaction in Excel * Send a confirmation email What previously required multiple people and several applications now happens automatically in seconds. Rather than automating individual tasks, Power Automate automates entire business processes. REAL-WORLD USE CASES Organizations use Power Automate across nearly every department. HR teams automate employee onboarding and vacation requests. Finance teams process invoice approvals and expense reports. Sales teams synchronize CRM information and customer notifications. IT departments automate user provisioning, ticket routing, and system monitoring. Managers receive scheduled reports without manually gathering data. Marketing teams connect forms, mailing lists, social media platforms, and approval workflows. The possibilities continue to expand as additional connectors become available.  Become a supporter of this podcast: https://www.spreaker.com/podcast/m365-fm-modern-work-security-and-productivity-with-microsoft-365--6704921/support [https://www.spreaker.com/podcast/m365-fm-modern-work-security-and-productivity-with-microsoft-365--6704921/support?utm_source=rss&utm_medium=rss&utm_campaign=rss].

AI Agents - Simply Explained

Welcome to another episode of Knowledge Nuggets with Mirko Peters. In this episode, we're exploring AI Agents—one of the fastest-growing concepts in artificial intelligence and the foundation of Microsoft's vision for the future of work. The term "AI agent" is often used interchangeably with chatbot, but they're fundamentally different. A chatbot answers questions. An AI agent pursues goals. It can reason, make decisions, interact with business systems, perform tasks autonomously, and report back when the work is complete. Throughout this episode, you'll learn what makes AI agents unique, how Microsoft enables them across Microsoft 365 and Azure, and which platform you should choose depending on your requirements. CHATBOTS VS. AI AGENTS The easiest way to understand AI agents is by comparing them with traditional chatbots. A chatbot is reactive. It waits for a user to ask a question, generates an answer, and the interaction ends. Its memory is usually limited to the current conversation, and it cannot perform work outside the chat interface. An AI agent works very differently. Instead of simply answering questions, an agent receives an objective and determines the necessary steps to accomplish it. It can access calendars, send emails, update SharePoint lists, create documents, interact with business systems, and complete workflows without continuous user supervision. Think of a chatbot as a receptionist answering questions at the front desk. An AI agent is more like a project manager who coordinates multiple tasks, communicates with different systems, and returns once the objective has been completed.  WHAT MAKES AN AI AGENT? An AI agent combines several capabilities that extend far beyond language generation. Modern agents can: * Reason through complex problems * Plan multiple steps * Use external tools * Access enterprise knowledge * Make decisions * Execute actions * Learn from context * Work autonomously toward a goal Rather than generating isolated responses, agents continuously evaluate what should happen next until the assigned task has been completed. This makes them ideal for business automation, customer service, IT operations, HR, procurement, finance, and countless other enterprise scenarios. MICROSOFT'S THREE AGENT BUILDING OPTIONS Microsoft provides three different approaches for building AI agents. Agent Builder Agent Builder is the simplest option. Included with Microsoft 365 Copilot, it enables users to create no-code agents simply by describing what they want in natural language. Knowledge can be sourced from: * SharePoint * OneDrive * Teams * Websites * Uploaded files * Organizational data This makes Agent Builder perfect for departmental knowledge assistants and internal Q&A solutions. Copilot Studio Copilot Studio introduces low-code capabilities. Developers and power users can connect over a thousand business systems through connectors, build custom conversation flows, trigger workflows automatically, and publish agents to Microsoft Teams, Microsoft 365 Copilot, or external websites. It's designed for organizations that want AI agents capable of interacting with real business processes instead of simply answering questions. Azure AI Foundry Azure AI Foundry provides complete developer control. Organizations can choose different AI models, build multi-agent architectures, customize orchestration logic, integrate advanced memory, and deploy enterprise-scale AI workloads running on managed Azure infrastructure. Foundry is the platform of choice for professional AI engineering teams building production-ready AI solutions. KNOWLEDGE, MEMORY, AND TOOLS An AI agent becomes valuable because it combines reasoning with enterprise knowledge. Depending on the platform, agents can retrieve information from: * Microsoft Graph * SharePoint * Teams * OneDrive * Dataverse * Outlook * External business systems * Custom APIs Unlike traditional AI models that rely only on their training data, enterprise agents continuously ground their responses using current organizational information. This dramatically improves accuracy while reducing hallucinations and ensuring answers remain relevant to the latest business data. DIGITAL IDENTITIES FOR AGENTS One of Microsoft's most significant innovations is giving AI agents their own digital identities. Each enterprise agent receives an identity within Microsoft Entra ID, similar to a human employee. This identity can include: * A mailbox * OneDrive storage * Teams presence * Security permissions * Authentication credentials Because agents authenticate through Microsoft Entra ID, they follow the same permission model as human users. If an agent isn't authorized to access a document, it cannot retrieve or use that information. This creates a secure foundation for enterprise AI adoption. WORK IQ AND ENTERPRISE GROUNDING Microsoft uses Work IQ to provide agents with organizational context. Instead of relying purely on general AI knowledge, Work IQ searches: * Emails * Documents * Teams conversations * Calendar events * SharePoint content * Microsoft Graph data Every query respects existing security permissions. Users only receive answers based on information they are already authorized to access, preserving enterprise security boundaries while significantly improving response quality. REAL-WORLD PROCUREMENT AGENT The episode demonstrates Microsoft's procurement agent as a practical example. The workflow begins with an approved agent published through an internal agent catalog. After receiving procurement policies, supplier information, and purchasing guidelines, the agent can independently: * Evaluate supplier options * Compare pricing * Review SLAs * Create purchase orders * Update SharePoint tracking systems * Report progress through Microsoft Teams Users simply assign the objective while the agent coordinates the individual tasks autonomously. This illustrates the transition from AI assistants toward digital coworkers capable of performing meaningful business work. Become a supporter of this podcast: https://www.spreaker.com/podcast/m365-fm-modern-work-security-and-productivity-with-microsoft-365--6704921/support [https://www.spreaker.com/podcast/m365-fm-modern-work-security-and-productivity-with-microsoft-365--6704921/support?utm_source=rss&utm_medium=rss&utm_campaign=rss].

From Data to Intelligent Agents: Building Trusted Enterprise AI with Microsoft AI Foundry with Shubhangi Goyal [MVP]

Enterprise AI is entering a new phase where success is no longer measured by impressive demos but by real business outcomes. Organizations are moving beyond experimenting with large language models and are now focused on building intelligent AI agents that are secure, scalable, and trusted by employees and customers alike. In this episode of the M365.fm podcast, Microsoft MVP Shubhangi Goyal joins Mirko Peters to explore how Microsoft AI Foundry enables enterprises to design, deploy, and govern AI solutions that create measurable value. WHY DATA QUALITY IS THE FOUNDATION OF EVERY SUCCESSFUL AI PROJECT One of the biggest takeaways from this conversation is that AI is only as good as the data behind it. Shubhangi explains why organizations must prioritize high-quality, well-governed data before investing heavily in AI initiatives. The discussion covers AI readiness, context engineering, responsible AI, and why clean enterprise data remains the most valuable asset for building reliable AI agents that users can actually trust.  MICROSOFT AI FOUNDRY EXPLAINED: FROM MODELS TO PRODUCTION-READY AI AGENTS Discover how Microsoft AI Foundry brings together foundation models, AI agents, evaluation tools, observability, guardrails, and enterprise governance into a single platform. Shubhangi explains how AI Foundry integrates with Microsoft Fabric, Azure AI Search, and enterprise knowledge sources while comparing its capabilities with Copilot Studio and discussing when each platform is the right choice for different AI scenarios.  BUILDING RESPONSIBLE AI WITH GOVERNANCE, SECURITY, AND TRUST Responsible AI is about much more than compliance. This episode explores AI governance frameworks, security controls, guardrails, hallucination mitigation, red teaming, explainable AI, and continuous evaluation. Learn why enterprises must embed governance into every stage of AI development instead of treating it as an afterthought and how trust ultimately determines whether an AI solution succeeds or fails. MAXIMIZING ROI FROM ENTERPRISE AI INVESTMENTS Many organizations invest heavily in AI but struggle to demonstrate measurable business value. Shubhangi shares practical advice on identifying high-impact AI use cases, selecting the right models, balancing cost versus performance, improving productivity, and calculating return on investment. She explains why successful AI adoption starts with solving real business problems instead of deploying AI simply because it's the latest trend.  THE FUTURE OF AGENTIC AI AND MICROSOFT'S AI ECOSYSTEM Looking ahead, the conversation explores the future of enterprise AI, including AI memory management, multi-agent systems, context engineering, enterprise knowledge, and the rapid evolution of Microsoft AI Foundry. Whether you're an IT professional, AI engineer, architect, data scientist, Microsoft partner, or business leader, this episode provides practical insights for building intelligent, trustworthy AI. WHAT YOU LEARN * What Microsoft AI Foundry is and when to use it * Why data quality is the foundation of successful AI * AI readiness strategies for enterprise organizations * AI governance, Responsible AI, and security best practices * Building intelligent AI agents with Microsoft technologies * Microsoft AI Foundry vs Copilot Studio vs Microsoft Fabric * Reducing hallucinations through evaluation and guardrails * Measuring ROI for enterprise AI investments * The future of agentic AI, memory management, and context engineering * Practical lessons from real-world enterprise AI projects  solutions that are ready for production. Become a supporter of this podcast: https://www.spreaker.com/podcast/m365-fm-modern-work-security-and-productivity-with-microsoft-365--6704921/support [https://www.spreaker.com/podcast/m365-fm-modern-work-security-and-productivity-with-microsoft-365--6704921/support?utm_source=rss&utm_medium=rss&utm_campaign=rss].