S2E01 Recap: Advanced Reasoning Session
About Model Mondays Want to know what Reasoning models are and how you can build advanced reasoning scenarios like a Deep Research agent using Azure AI Foundry? Check out this recap from Model Mondays Season 2 Ep 1. Model Mondays is a weekly series to help you build your model IQ in three steps: 1. Catch the 5-min Highlights on Monday, to get up to speed on model news 2. Catch the 15-min Spotlight on Monday, for a deep-dive into a model or tool 3. Catch the 30-min AMA on Friday, for a Q&A session with subject matter experts Want to follow along? Register Here- to watch upcoming livestreams for Season 2 Visit The Forum- to see the full AMA schedule for Season 2 Register Here - to join the AMA on Friday Jun 20 Spotlight On: Advanced Reasoning This week, the Model Mondays spotlight was on Advanced Reasoning with subject matter expert Marlene Mhangami. In this blog post, I'll talk about my five takeaways from this episode: Why Are Reasoning Models Important? What Is an Advanced Reasoning Scenario? How Can I Get Started with Reasoning Models ? Spotlight: My Aha Moment Highlights: What’s New in Azure AI 1. Why Are Reasoning Models Important? In today's fast-evolving AI landscape, it's no longer enough for models to just complete text or summarize content. We need AI that can: Understand multi-step tasks Make decisions based on logic Plan sequences of actions or queries Connect context across turns Reasoning models are large language models (LLMs) trained with reinforcement learning techniques to "think" before they answer. Rather than simply generating a response based on probability, these models follow an internal thought process producing a chain of reasoning before responding. This makes them ideal for complex problem-solving tasks. And they’re the foundation of building intelligent, context-aware agents. They enable next-gen AI workflows in everything from customer support to legal research and healthcare diagnostics. Reason: They allow AI to go beyond surface-level response and deliver solutions that reflect understanding, not just language patterning. 2. What does Advanced Reasoning involve? An advanced reasoning scenario is one where a model: Breaks a complex prompt into smaller steps Retrieves relevant external data Uses logic to connect dots Outputs a structured, reasoned answer Example: A user asks: What are the financial and operational risks of expanding a startup to Southeast Asia in 2025? This is the kind of question that requires extensive research and analysis. A reasoning model might tackle this by: Retrieving reports on Southeast Asia market conditions Breaking down risks into financial, political, and operational buckets Cross-referencing data with recent trends Returning a reasoned, multi-part answer 3. How Can I Get Started with Reasoning Models? To get started, you need to visit a catalog that has examples of these models. Try the GitHub Models Marketplace and look for the reasoning category in the filter. Try the Azure AI Foundry model catalog and look for reasoning models by name. Example: The o-series of models from Azure Open AI The DeepSeek-R1 models The Grok 3 models The Phi-4 reasoning models Next, you can use SDKs or Playground for exploring the model capabiliies. 1. Try Lab 331 - for a beginner-friendly guide. 2. Try Lab 333 - for an advanced project. 3. Try the GitHub Model Playground - to compare reasoning and GPT models. 4. Try the Deep Research Agent using LangChain - sample as a great starting project. Have questions or comments? Join the Friday AMA on Azure AI Foundry Discord: 4. Spotlight: My Aha Moment Before this session, I thought reasoning meant longer or more detailed responses. But this session helped me realize that reasoning means structured thinking — models now plan, retrieve, and respond with logic. This inspired me to think about building AI agents that go beyond chat and actually assist users like a teammate. It also made me want to dive deeper into LangChain + Azure AI workflows to build mini-agents for real-world use. 5. Highlights: What’s New in Azure AI Here’s what’s new in the Azure AI Foundry: Direct From Azure Models - Try hosted models like OpenAI GPT on PTU plans SORA Video Playground - Generate video from prompts via SORA models Grok 3 Models - Now available for secure, scalable LLM experiences DeepSeek R1-0528 - A reasoning-optimized, Microsoft-tuned open-source model These are all available in the Azure Model Catalog and can be tried with your Azure account. Did You Know? Your first step is to find the right model for your task. But what if you could have the model automatically selected for you_ based on the prompt you provide? That's the magic of Model Router a deployable AI chat model that dynamically selects the best LLM based on your prompt. Instead of choosing one model manually, the Router makes that choice in real time. Currently, this works with a fixed set of Azure OpenAI models, including a reasoning model option. Keep an eye on the documentation for more updates. Why it’s powerful: Saves cost by switching between models based on complexity Optimizes performance by selecting the right model for the task Lets you test and compare model outputs quickly Try it out in Azure AI Foundry or read more in the Model Catalog Coming Up Next Next week, we dive into Model Context Protocol, an open protocol that empowers agentic AI applications by making it easier to discover and integrate knowledge and action tools with your model choices. Register Here to get reminded - and join us live on Monday! Join The Community Great devs don't build alone! In a fast-pased developer ecosystem, there's no time to hunt for help. That's why we have the Azure AI Developer Community. Join us today and let's journey together! Join the Discord - for real-time chats, events & learning Explore the Forum - for AMA recaps, Q&A, and help! About Me. I'm Sharda, a Gold Microsoft Learn Student Ambassador interested in cloud and AI. Find me on Github, Dev.to,, Tech Community and Linkedin. In this blog series I have summarizef my takeaways from this week's Model Mondays livestream .Getting Started with the AI Toolkit: A Beginner’s Guide with Demos and Resources
If you're curious about building AI solutions but don’t know where to start, Microsoft’s AI Toolkit is a great place to begin. Whether you’re a student, developer, or just someone exploring AI for the first time, this toolkit helps you build real-world solutions using Microsoft’s powerful AI services. In this blog, I’ll Walk you through what the AI Toolkit is, how you can get started, and where you can find helpful demos and ready-to-use code samples. What is the AI Toolkit? The AI Toolkit is a collection of tools, templates, and sample apps that make it easier to build AI-powered applications and copilots using Microsoft Azure. With the AI Toolkit, you can: Build intelligent apps without needing deep AI expertise. Use templates and guides that show you how everything works. Quickly prototype and deploy apps with natural language, speech, search, and more. Watch the AI Toolkit in Action Microsoft has created a video playlist that covers the AI Toolkit and shows you how to build apps step-by-step. You can watch the full playlist here: It is especially useful for developers who want to bring AI into their projects, but also for beginners who want to learn by doing. AI Toolkit Playlist – https://aka.ms/AIToolkit/videos These videos help you understand the flow of building AI agents, using Azure OpenAI, and other cognitive services in a hands-on way. Explore Sample Projects on GitHub Microsoft also provides a public GitHub repository where you can find real code examples built using the AI Toolkit. Here’s the GitHub repo: AI Toolkit Samples – https://github.com/Azure-Samples/AI_Toolkit_Samples This repository includes: Sample apps using Azure AI services like OpenAI, Cognitive Search, and Speech. Instructions to deploy apps using Azure. Code that you can clone, test, and build on top of. You don’t have to start from scratch just open the code, understand the structure, and make small edits to experiment. How to Get Started Here’s a simple path if you’re just starting: Watch 2 or 3 videos from the AI Toolkit Playlist. Go to the GitHub repository and try running one of the examples. Make small changes to the code (like updating the prompt or output). Try deploying the solution on Azure by following the guide in the repo. Keep building and learning. Why This Toolkit is Worth Exploring As someone who is also learning and experimenting, I found this toolkit to be: Easy to understand, even for beginners. Focused on real-world applications, not just theory. Helpful for building responsible AI solutions with good documentation. It gives a complete picture — from writing code to deploying apps. Final Thoughts The AI Toolkit helps you start your journey in AI without feeling overwhelmed. It provides real code, real use cases, and practical demos. With the support of Microsoft Learn and Azure samples, you can go from learning to building in no time. If you’re serious about building with AI, this is a resource worth exploring. Continue the discussion in the Azure AI Foundry Discord community at Https://aka.ms/AI/discord Join the Azure AI Foundry Discord Server! References AI Toolkit Playlist (YouTube) https://aka.ms/AIToolkit/videos AI Toolkit GitHub Repository https://github.com/Azure-Samples/AI_Toolkit_Samples Microsoft Learn: AI Toolkit Documentation https://learn.microsoft.com/en-us/azure/ai-services/toolkit/ Azure AI Services https://azure.microsoft.com/en-us/products/ai-services/
M365 Developers Update | June 2025
Spotlight Bring the AI capabilities of Copilot into your apps and agents with Microsoft 365 Copilot APIs Learn More Dive deeper into the latest wave of updates to Copilot Studio, now with powerful new tools for multi-agent systems, enterprise data access, custom AI tuning, and more. See all updates Explore the Microsoft 365 Agents Toolkit, an evolution of Microsoft Teams Toolkit, designed to help you build agents and apps for Microsoft 365 Copilot and beyond. Read the docs Introducing the Agent Store: A centralized, curated marketplace within Microsoft 365 Copilot to browse, install, and try agents tailored to your needs. Start browsing Learn Use the Microsoft 365 Agents SDK to build agents for Microsoft 365 Copilot, Microsoft Teams, and more using models provided by Azure AI Foundry. Watch how Try the latest version of Dev Proxy, now with the capability to help you understand language models’ usage and costs in your applications. Get the latest Learn more about Athena, an AI-powered collaborative agent, how it was built, and how to create your own version of Athena right within Microsoft Teams. Read now Keep up to date On Demand: Catch up on all of our breakout sessions from Build Get started YouTube: Tune in for tutorials on the latest dev tools Subscribe now LinkedIn: Get the latest news, product announcements, demos, and more Follow us Community Calls: Learn from our experts on a variety of Microsoft 365 platform topics Join a callMVP Collective Launches In-Depth Guide on SharePoint Content AI
MVP and Regional Director Gokan Ozcifci, together with eight fellow Microsoft MVPs, has co-authored SharePoint Content AI, Solutions and Advanced Administration - a new book that delves into the intersection of artificial intelligence and SharePoint. We spoke with the authors to learn more about their collaboration, what inspired the project, and the key themes they explore. What inspired you to collaborate on this eBook, and how did the idea for focusing on AI and SharePoint come about? Gokan Ozcifci, Belgium: The inspiration behind this book stems from observing how rapidly AI is transforming the way organizations manage content, particularly within Microsoft 365 and SharePoint. As this transformation accelerated, I noticed a growing gap: many industry experts, digital transformation directors, and business leaders were eager to embrace these AI-driven tools but found it difficult to understand how capabilities like Autofill, AI-powered metadata, OCR, or governance solutions translate into real-world value. That’s when a group of SharePoint enthusiasts, MVPs, consultants, and practice owners joined forces with a shared goal—to bridge that gap. We set out to create more than just a technical guide. We aimed to build a resource grounded in practical experience, offering clear explanations and actionable insights for those navigating the evolving world of Content AI. SharePoint naturally became our focal point due to its central role in enterprise content management and its rapid evolution in tandem with Microsoft’s AI strategy. Our goal was to demystify the technology, understand the requirements, translate the needs, and show how AI can empower organizations to manage content more intelligently, securely, and efficiently. Can you share a real-world example where AI significantly enhanced SharePoint content management or administration? Frane Borozan – MVP, Croatia: A global enterprise had a use case to classify thousands of legacy contracts in SharePoint. SharePoint content AI extracts metadata, such as expiration dates, the names of signatories, and the validity period of the contract, as well as similar information that is available within the contract. This cut manual effort by almost 90% to hire the workforce to review all these legacy contracts. Noorez Khamis – MVP, Canada: For one client they previously had 5–10 interns manually logging into 20 banking websites each month to download client statements, upload them to SharePoint, tag metadata, and update Salesforce—an inefficient and error-prone process. Now, an automated solution using Power Automate Desktop Flows handles document retrieval, Syntex extracts key metadata, and a validation Power App ensures data accuracy before integrating with Salesforce for approvals and updates. This end-to-end system eliminates manual effort, increases accuracy, and streamlines document processing across platforms. Drew Madelung – MVP, USA: As an M365 consultant, I work with multiple customers and write unique and complex statements of work. I actively utilize SharePoint Content AI features, such as autofill columns, to help summarize and provide key metadata from statements of work, enabling me to discover details from prior and existing projects where overlap is likely to occur. Mike Maadarani – MVP, Canada: AI was deployed at a university to manage their application admissions process. The Content AI significantly improved the classification and extraction of information from various application formats. This process reduced manual work by over 98%, resulting in substantial savings and a high return on investment for the client. Antonio Maio – MVP, Canada: I had a client who greatly benefited from Content AI models in SharePoint Online. They’re a corporate real estate firm that utilizes Content AI models to process lease agreements and rental contracts, automatically extracting key metadata values from these content types. This metadata automatically populates columns in SharePoint libraries, which then drives business process automation and retention policies for those documents. This all happens by users simply uploading new documents into SharePoint libraries. How do you see the role of AI evolving in the SharePoint ecosystem over the next few years? Frane Borozan – MVP, Croatia: SharePoint, as the content management platform, doesn't have a future without the help of AI. Use cases are varied, ranging from extracting metadata from content to helping create new content. I believe that with the help of AI, the possibilities of SharePoint are unlimited. Noorez Khamis – MVP, Canada: Copilot is making SharePoint the go-to content management system by transforming how you discover, create, and interact with content. You can now ask for what you need, generate pages from your existing documents, and get personalized answers through AI-powered SharePoint Agents. With more intelligent automation, beautiful intranet design, and fewer clicks, SharePoint feels more like an intelligent assistant than a static site. Vlad Catrinescu – MVP, Canada: AI will continue transforming how we work, and SharePoint is no exception. Today, we’re already seeing AI help fill in metadata for document libraries. But imagine if AI could go further: automatically suggest and create the right columns, build content types based on the documents you upload, or even configure web parts through natural language prompts. SharePoint has always been a powerful platform, but it hasn’t always been the easiest to use. AI has the potential to make that power more accessible to every user, not just the experts. Drew Madelung – MVP, USA: I would like to see SharePoint Content AI evolve in a way that identifies opportunities for AI using existing content that automatically configures it without user intervention to improve discoverability. The ability to configure and work with these AI features should have a minimal learning curve and be integrated seamlessly, without requiring specialized technical skills. Mike Maadarani – MVP, Canada: As artificial intelligence continues to advance, it is anticipated that content management will become fully automated, reducing the need for administrators to establish and enforce rules. AI algorithms will be significantly more sophisticated, enhancing their ability to comprehend an organization's policies, the nature of the content being added, and the necessary actions required by the rules. Antonio Maio – MVP, Canada: I think auto-fill columns will have a significant impact on SharePoint Online. Metadata is a core element of good information management, but we know that users don’t want to fill in metadata. They’re busy and often move too quickly from task to task, leaving little time to provide a wealth of metadata elements. SharePoint’s auto-fill columns offer an easy way for us to automatically extract metadata, based on a prompt that’s supplied to the column. Joanne Klein – MVP, Canada: As a security and compliance professional, I observe the evolution of AI data governance, which aims to control the proliferation, access, and lifecycle of AI solutions across SharePoint within the enterprise. We need to elevate AI to be a core pillar within an organization's holistic data governance strategy. What advice would you give to SharePoint professionals who are just beginning to explore AI-powered capabilities? Frane Borozan – MVP, Croatia: If you're starting with AI in SharePoint, taxonomy tagging is a perfect first step. It shows how AI can reduce manual effort and bring structure to your content. Set up a managed metadata column linked to your term store and let AI handle tagging based on document content. It’s a simple way to improve search, consistency, and governance—without heavy customization. Start here, learn the basics, and expand as you go. Noorez Khamis – MVP, Canada: Start with the basics and learn how to prompt effectively while using the full Microsoft 365 Copilot capabilities across all the workplace tools you use every day, such as Teams, SharePoint, Word, Excel, PowerPoint, and Outlook. Take the time to revise your prompts and use templates that have been proven to work, saving you time, streamlining tasks, and boosting productivity. In SharePoint specifically, explore how to create pages, rewrite content, and use AI-powered SharePoint agents. Vlad Catrinescu – MVP, Canada: Get hands-on as early as possible. Theory is great, but real understanding comes from testing in your environment. Set up a lab, start small, and explore practical use cases where AI can help automate or enhance existing processes. With the Pay-As-You-Go model, there’s no upfront cost—you only pay for what you use. That said, I highly recommend setting a budget cap in Azure to avoid surprises. Drew Madelung – MVP, USA: SharePoint remains essentially unchanged in many ways, and it is still essential to understand the concepts of content types, columns, and permission hierarchies to implement advanced AI solutions against your organization's content effectively. Mike Maadarani – MVP, Canada: IT professionals must stay current with evolving technologies, particularly in the rapidly changing field of AI. With the emergence of AI agents, I recommend acquiring skills in creating, managing, and deploying agents within Microsoft 365 to enhance the integration and utilization of AI in their organizations. Antonio Maio – MVP, Canada: Be curious about AI - play with the AI technology that’s built into SharePoint; experiment with it to see what best benefits your organization to improve how you specifically manage information. Your experience will be different than everyone else’s, so try different things to figure out what works for you and your users. Joanne Klein – MVP, Canada: If your SharePoint setup is a mess, your AI will be too. Solid, well-defined structure and smart governance (site owner stewardship, explicit permissions, retention/deletion to clean up ROT, and data protection controls) are like laying concrete before building—skip it, and your AI’s standing on quicksand. Access the full 194-page e-book, SharePoint Content AI, Solutions and Advanced Administration, at the following link SharePoint Content AI, Solutions and Advanced AdministrationPartner Case Study | SOUTHWORKS
Each year, the NFL Combine brings together coaches and scouts from its 32 clubs to evaluate the talent and performance data of more than 300 of the best athletes in the world. The decisions made at the Combine shape the future of football clubs and players alike. Delays or oversights during the event can come with a cost. The need to speed access to data and insights within the existing NFL Combine App drove the latest evolution of the app, which has supported clubs at Combine for several seasons. The addition of an AI assistant, designed to allow scouts and coaches to get instantaneous player insights and ask their questions with common Combine vernacular, marked a major leap forward. “We went for it! We decided to dramatically change the game for these clubs by giving them exactly what they need when they need it,” explains Jeff Foster, President of National Football Scouting (NFS). “We wanted to transform hours of tedious, manual effort into seconds of grab-and-go data.” Continue reading here Explore all case studies or submit your own Subscribe to case studies tag to follow all new case study posts. Don't forget to follow this blog to receive email notifications of new stories!
🚀 Introdução à IA e ao MS Copilot – Participe da Sessão ao Vivo! ✨
👋 Quer explorar IA e o Microsoft Copilot de forma prática para o ensino? Participe da sessão “Introdução à IA e ao MS Copilot”, pensada especialmente para docentes (participantes ou não do MSLE). Vamos aprender os fundamentos da IA generativa, como usar prompts e aplicar essas ferramentas em sala de aula. 📌 Sessão em português, com exemplos práticos, materiais prontos para uso e um espaço acolhedor para tirar dúvidas. 👉 Para acessar diretamente esta sessão na data e horário, clique aqui: Participar da sessão no horário. 🧭 Confira a aba de eventos aqui na home da nossa comunidade para ver a agenda com todas as sessões disponíveis. Vamos aprender e evoluir juntos! Atenciosamente, Henoc Freire Gerente de Comunidade
AI in IDD
Here is a little bout how I am using AI in IDD. Our licensed Co-Pilot, has proven to be incredibly productive and time-saving, especially in reviewing documents for errors against regulations, identifying discrepancies between various care plans for the same individual, analyzing and summarizing documents for relevant information in investigations (reducing the closure time from 90 days to 7), suggesting corrective actions for internal/external audits, trending data sets (such as surveys, assessments, and audits), and proposing opportunities and solutions in light of the regulations. AI can audit documents, policies, and procedures with extreme accuracy against 6400 and 6100 regulations (PA regs). We have recently implemented this solution to streamline and audit our Medicaid billing process prior to submission. It enables us to identify billing errors, uncover unbilled days we should be billing for, and flag days we should not be billing - significantly reducing the time required for this process from 30 hours to 4 for a months' worth of billing. For systems that are not directly compatible with AI integration, we leverage AI to develop macros that process raw data exports. This allows us to extract precisely the information we need in seconds, rather than spending hours on manual analysis.
