Episodes

AI agents are often managed like traditional service accounts—but that mindset creates serious security and governance risks. This episode explains why AI agents should be treated as autonomous digital identities with their own lifecycle, permissions, and accountability instead of static technical accounts. The discussion highlights that agents don't just authenticate to services—they make decisions, access business data, trigger workflows, and interact with multiple systems. Because of this, identity becomes the primary security boundary. Organizations should assign every agent its own identity, apply least-privilege access, enforce Zero Trust principles, and continuously monitor behavior rather than relying on broad, long-lived permissions. The episode also covers the importance of lifecycle management for non-human identities. Just like employees, AI agents require onboarding, role assignments, permission reviews, auditing, and secure decommissioning. Without proper governanc…

Work IQ represents Microsoft’s next major evolution for Microsoft 365 Copilot, transforming it from a simple AI assistant into an intelligence layer that understands people, teams, projects, and organizational knowledge. Instead of relying only on documents, Work IQ builds a contextual understanding of how work happens across Microsoft 365, enabling more accurate answers, better recommendations, and smarter AI agents. In this episode, we explore how Work IQ combines signals from emails, meetings, chats, documents, calendars, organizational structures, and business processes to create a rich knowledge graph that powers Microsoft 365 Copilot. The discussion explains why context has become the most valuable asset in enterprise AI and how organizations can improve productivity without constantly training custom AI models. The episode also covers the architectural foundations behind Work IQ, its role in multi-agent collaboration, security boundaries, governance, and how it helps redu…

Building AI agents is no longer just a development challenge—it’s an operational one. In this episode of the M365 FM Podcast, we explore the concept of the Synthetic Platform Team and how organizations can successfully operationalize Azure Copilot Agents at enterprise scale. Rather than treating AI agents as isolated applications, the discussion introduces a platform engineering approach where reusable infrastructure, governance, observability, and automation become the foundation for every agent. The episode explains how platform teams can standardize identity, security, model access, prompt management, deployment pipelines, monitoring, and cost optimization while giving development teams the freedom to build innovative AI solutions. Topics include Azure AI Foundry, Azure OpenAI, Copilot technologies, Infrastructure as Code, CI/CD, telemetry, policy enforcement, and lifecycle management for AI agents. Listeners will learn why successful enterprise AI depends less on individual …

Custom integrations have long been one of the biggest bottlenecks in enterprise applications, but Dataverse MCP is changing that. In this episode of the M365.fm podcast, Mirko Peters is joined by Microsoft MVP Nathan Rose to explore how the Model Context Protocol (MCP) and Dataverse are transforming the way AI agents connect with business data. The discussion explains why traditional APIs, custom connectors, and one-off integrations struggle to scale in an AI-first world. Instead, Dataverse MCP introduces a standardized, secure, and context-aware approach that allows Microsoft Copilot Studio and AI agents to access enterprise data with far less custom development. Nathan shares practical insights into MCP architecture, governance, security, and how organizations can build reusable integrations instead of maintaining complex point-to-point connections. The conversation also covers the future of agentic AI, Microsoft's vision for interoperable AI systems, and what developers, arch…

The software development workflow is undergoing its biggest transformation since the rise of modern IDEs. In this episode, we explore how AI is turning the terminal from a simple command-line interface into the orchestration layer of modern software engineering. Rather than manually executing commands, developers increasingly define goals while intelligent agents plan, execute, validate, and refine the work. We explain why AI-assisted coding alone isn't enough to solve today's productivity challenges and examine the hidden bottlenecks in code reviews, CI/CD pipelines, modernization projects, and deployment workflows. The discussion highlights the shift from traditional tools that respond to commands to autonomous agents that understand intent, make decisions, and operate within defined policies. A central focus is the Agentic Developer Stack, built on four key layers: orchestration, transformation, validation, and execution. Together, these layers enable AI agents to modernize l…

In this episode of the M365 FM podcast, Sailaja Mantripragada explores how organizations can move beyond simple AI assistants and build enterprise-grade AI agents using Microsoft Copilot Studio and the Power Platform. The conversation focuses on creating intelligent, business-driven agents that deliver measurable value while remaining secure, compliant, and governable. Sailaja shares practical insights from real-world enterprise projects, explaining how organizations can design AI agents that integrate with business processes, automate complex workflows, and leverage organizational knowledge through Microsoft 365, Dataverse, and Power Platform services. The discussion highlights the importance of balancing innovation with governance, ensuring that AI solutions remain aligned with security, compliance, and responsible AI principles. A major theme of the episode is AI governance. Sailaja discusses why governance must be built into AI initiatives from the beginning rather than adde…

In this episode of the M365.FM Podcast, we explore why mastering Dataverse business skills is becoming one of the most important capabilities for organizations building scalable solutions on Microsoft Power Platform. The discussion goes far beyond tables and columns, focusing on how Dataverse can be used to capture business knowledge, processes, and operational intelligence in a way that supports long-term growth and automation. You’ll learn why successful Power Apps, Power Automate solutions, and AI-powered business processes depend on strong data modeling foundations rather than simply creating forms, workflows, or user interfaces. The episode highlights how well-designed Dataverse structures help organizations maintain consistency, improve governance, and reduce technical debt as applications scale across departments and business units. The conversation also examines the emerging role of Dataverse business skills, showing how organizations can transform business processes int…

In this episode of the M365.fm Podcast, Mirko Peters explores whether custom APIs are becoming obsolete and how Microsoft’s new Rayfin platform could fundamentally change the way enterprise applications are built. Traditionally, developers have spent countless hours creating backend services, authentication layers, databases, APIs, and infrastructure before delivering actual business value. Rayfin introduces a different approach by offering a Backend-as-a-Service (BaaS) model that automates much of this complexity. The episode examines how Rayfin enables developers to define data models, business logic, security policies, and APIs directly in code while Microsoft Fabric automatically provisions and manages the backend infrastructure. Instead of building and maintaining custom APIs from scratch, organizations can focus on application functionality while inheriting enterprise-grade governance, security, compliance, and scalability from the Fabric platform. Mirko discusses the broa…

In this episode of the M365 Show, Mirko Peters sits down with Sandra Kiel to explore a fascinating question: what can enterprise software learn from video games? Drawing on her unique background in gaming, user experience, and Microsoft technologies, Sandra explains why many business applications struggle with adoption while games keep millions of people engaged for hours. The conversation dives into the psychology behind successful game design and how those same principles can transform workplace software. Sandra discusses concepts such as clear goals, instant feedback, progression systems, and user motivation. Rather than focusing on gamification as points and badges, she emphasizes designing experiences that make users feel confident, capable, and motivated to continue. Mirko and Sandra examine why many enterprise tools overwhelm users with complexity and how game-inspired design patterns can improve onboarding, learning, and long-term adoption. They also discuss the growing …

The traditional SharePoint intranet is reaching the end of its lifecycle. For years, organizations have relied on manually designed sites, static navigation structures, and carefully curated content pages that quickly become outdated. In this episode of the M365 FM Podcast, we explore how artificial intelligence is transforming intranet design from a static publishing model into a dynamic, adaptive experience. The discussion examines why traditional intranets struggle to keep pace with modern workplaces. Employees expect personalized experiences, relevant information, and instant access to knowledge, yet many intranets still rely on manual content management and one-size-fits-all information architectures. As AI capabilities mature, this approach becomes increasingly difficult to justify. A central theme of the episode is that AI will not simply enhance intranets—it will fundamentally redesign how they are built and maintained. Instead of administrators manually creating navigat…

Most organizations approach enterprise AI with a simple vision: one Copilot, one interface, and one large language model capable of handling every business task. In this episode of the M365 FM Podcast, we challenge that assumption and explore why the future of enterprise AI belongs to a “Mixture of Experts” architecture rather than a single general-purpose assistant. As AI adoption expands across the enterprise, a single bot is often expected to perform knowledge retrieval, policy interpretation, workflow automation, document summarization, data extraction, and countless other responsibilities. While this sounds efficient on paper, it frequently creates hidden costs, inconsistent results, governance challenges, and growing operational complexity. The episode explains why specialized AI agents deliver better outcomes than one generalist model trying to do everything. Instead of relying on a single intelligence layer, organizations can build networks of expert agents that focus on…

Hybrid work has fundamentally changed how organizations connect employees, but creating engaging virtual events remains a major challenge. In this episode of the M365 FM Podcast, we explore what it really takes to engineer successful high-stakes hybrid events using Microsoft Teams Immersive Spaces and Microsoft Mesh. Rather than focusing on product marketing, the discussion examines the technical realities that determine whether an immersive event succeeds or fails. A key concept introduced is the “Ghost Town Effect” — when participants abandon virtual experiences due to poor performance, lagging avatars, broken spatial audio, synchronization issues, and low engagement. Even the most visually impressive virtual environments can quickly become ineffective if the user experience suffers. The episode explores the evolution of Microsoft Mesh and Teams Immersive Spaces, highlighting how organizations can move beyond traditional video meetings to create more interactive and collaborat…

veryone is talking about Private RAG, sovereign AI, regional hosting, and keeping enterprise data inside controlled environments. But in this episode of M365 FM, Mirko Peters explores a critical security gap that many organizations overlook: data sovereignty is not the same as data security. The episode examines what happens when documents leave systems like SharePoint and Microsoft 365 and are ingested into vector databases for Retrieval-Augmented Generation (RAG) solutions. While organizations often focus on where data is stored, they frequently ignore what happens to permissions, access controls, and authorization models during the indexing process. The result can be a highly capable AI system that unintentionally exposes sensitive information to users who should never have access to it. A major focus is the concept of authorization-aware retrieval. Listeners learn why self-hosting, VPN access, or private infrastructure alone do not guarantee security. The episode breaks down…

SharePoint has become the central repository for business knowledge, documents, contracts, policies, and operational records. Yet many organizations are sitting on a hidden problem: their content lacks the metadata needed to make that information truly discoverable, governable, and AI-ready. In this episode of M365 FM, Mirko Peters explores the growing “metadata gap” and why it has become one of the biggest risks in modern Microsoft 365 environments. While organizations invest heavily in SharePoint, Copilot, search, and automation, many still rely on inconsistent folder structures, poor document classification, and manual filing processes that create data chaos over time. The discussion explains how missing or inconsistent metadata impacts far more than search. It affects compliance, records management, retention policies, security controls, business process automation, and the quality of AI-generated results. As Microsoft Copilot and other AI services depend on context to under…

In this episode of M365 FM, Mirko Peters sits down with Microsoft MVP Jonathan Hope to explore why identity has become the most critical security boundary in modern cloud environments. As organizations move deeper into Microsoft 365, Azure, and cloud-native architectures, traditional network-based security models are no longer enough. Jonathan explains how Microsoft Entra ID acts as the control plane for security and why mismanaged identities, excessive permissions, and poorly maintained Conditional Access policies create significant risk. The conversation dives into the concept of “identity debt” — the gradual accumulation of exceptions, legacy configurations, guest accounts, workload identities, and hybrid synchronization issues that weaken an organization’s security posture over time. The episode covers practical strategies for implementing Zero Trust principles, designing effective Conditional Access policies, enforcing least-privilege access, and protecting privileged accou…

Artificial Intelligence is transforming the way organizations manage knowledge, documents, and collaboration. Yet as companies rush to adopt AI assistants and large language models, one question continues to dominate every conversation: how can you benefit from AI without exposing sensitive business information to external services? In this episode of M365 FM, Mirko Peters explores how organizations can run Local Llama models directly against SharePoint content while maintaining complete control over their data. Instead of sending confidential documents, intellectual property, customer information, and internal knowledge to cloud-hosted AI platforms, organizations can build private AI solutions that keep processing inside their own environment. The episode breaks down the architecture behind local AI deployments, explaining how open-source LLMs, retrieval-augmented generation (RAG), semantic search, and document embeddings can be combined with SharePoint to create intelligent kn…

Artificial Intelligence is reshaping the workplace faster than ever, creating both new opportunities and new challenges for professionals across every industry. In this episode of M365 FM, Mirko Peters sits down with Sarah Jones, technology recruiter, career coach, freelancer, and community advocate, to discuss how individuals can futureproof their careers in an AI-driven world. Drawing on more than 20 years of experience in recruitment and career development, Sarah explains why technical AI skills are becoming increasingly important, while human-centered capabilities such as communication, leadership, trust-building, and relationship management are becoming even more valuable. As automation takes over repetitive work, the ability to collaborate, influence, and connect with people will remain a key competitive advantage. The conversation explores how AI is transforming hiring practices, recruitment processes, and workplace expectations. Sarah shares practical insights into appli…

AI agents are only as valuable as the systems they can access. While much of the industry focuses on models, prompts, and reasoning capabilities, the real challenge in enterprise AI is connectivity. In this episode, we take a deep architectural dive into the Model Context Protocol (MCP), the emerging standard that is rapidly becoming the integration layer for Microsoft Copilot, custom AI agents, Dynamics 365, Azure services, and enterprise applications. The discussion explores why traditional APIs were designed for developers and applications, not autonomous AI agents that need to dynamically discover capabilities, understand available tools, and execute actions across business systems. As organizations deploy more AI solutions, integration has become one of the biggest bottlenecks to scaling enterprise AI. MCP addresses this challenge by providing a standardized protocol that allows agents to interact with tools, resources, and external systems in a more intelligent and flexible w…

Microsoft Project Online is approaching retirement, and organizations are facing a critical decision: simply replace an aging project management tool or use the opportunity to modernize how projects are delivered across the business. In this episode, Joe Griffin, Microsoft MVP, CEO of proMX UK, and one of the leading experts in Dynamics 365 Project Operations, explores the future of project management within the Microsoft ecosystem. The conversation examines why Dynamics 365 Project Operations is becoming the strategic successor for many project-driven organizations. Unlike traditional project management solutions focused primarily on schedules and tasks, Project Operations combines project planning, resource management, budgeting, financial tracking, time and expense management, invoicing, and AI-powered insights within a single platform built on Microsoft Dataverse. Joe shares practical guidance for organizations preparing for the retirement of Project Online, including migrat…

Most organizations believe hallucinations are the biggest risk in enterprise AI. In reality, one of the most dangerous threats is something far less visible: Indirect Prompt Injection. In this episode, we explore how trusted documents, emails, SharePoint content, Teams conversations, and knowledge bases can become attack vectors that manipulate AI systems without ever compromising the underlying infrastructure. The episode examines why Retrieval-Augmented Generation (RAG), the foundation behind Microsoft 365 Copilot, Azure AI Foundry solutions, and many enterprise AI assistants, introduces an entirely new security challenge. Unlike traditional software, large language models cannot reliably separate data from instructions. Every piece of retrieved content becomes part of the model's context, allowing hidden commands, poisoned documents, metadata, and embedded instructions to influence AI behavior. Listeners will learn how indirect prompt injection works, why system prompts are n…

Most organizations believe they are building AI agents. In reality, many are still deploying advanced chatbots that operate within isolated applications. These systems can answer questions and generate content, but they lack persistence, long-term memory, and the ability to coordinate work across enterprise systems. In this episode, we explore one of the most important shifts happening in enterprise AI: the evolution from standalone conversational experiences to connected agent ecosystems. Instead of treating AI as a feature inside Microsoft Teams, Slack, or individual business applications, organizations are beginning to build persistent agent fabrics that can collaborate, share context, and execute tasks across multiple platforms. We discuss why memory, orchestration, identity, and governance are becoming critical components of modern AI architectures. The conversation examines how enterprises are moving beyond simple prompt-response interactions toward systems where AI agents…

In this episode, we explore why many organizations are making a critical mistake when building AI solutions: creating agents that operate in isolation. While individual agents can be powerful, they often become disconnected silos that lack the ability to coordinate across systems, processes, and business functions. The conversation focuses on Azure Logic Apps as the “nervous system” for enterprise AI, providing the orchestration layer that connects agents, applications, and workflows. Rather than viewing agents as standalone tools, organizations should design them as part of a larger ecosystem where events, messages, and automated processes enable collaboration and intelligent decision-making. We discuss the principles of event-driven architecture, the role of integration in modern AI systems, and how Logic Apps can connect Microsoft 365, Azure services, business applications, and external platforms. The episode also covers governance, scalability, and operational visibility, sh…

Multi-Agent Systems, Copilot Studio, Microsoft Agent Framework, Intelligent Automation, AI Orchestration, Microsoft Foundry, Artificial Intelligence, David Lorenzo Lopez, MC65 Podcast, Generative AI, Autonomous Agents, Enterprise AI, Microsoft Azure, Microservices Architecture, LLM Agents, AI for Business, GPT-3.5, Software Development, Microsoft Ecosystem

The rise of enterprise AI has created a fundamental challenge: how can organizations leverage powerful language models without exposing their most valuable proprietary data? In this episode of M365.fm, we explore the growing adoption of Private LoRA (Low-Rank Adaptation) as a practical architecture for secure AI systems built on sensitive enterprise information. Rather than sending confidential documents, intellectual property, customer records, or internal knowledge to public AI services, Private LoRA enables organizations to adapt and customize foundation models while keeping their data within controlled environments. The discussion explains why traditional fine-tuning approaches are often expensive, difficult to govern, and introduce significant security and compliance concerns. The episode breaks down how LoRA works by modifying only a small subset of model parameters, allowing organizations to create specialized AI capabilities without retraining entire large language model…