Uploaded October 2026 | Updated November 2026, 7 hours ago
In this webinar, you'll learn what Infobip and Microsoft have prepared to make your ACS migration smooth - from signup and onboarding with Infobip, to configuring your connection points back to your Azure solution, all the way to sending your first message through Infobip instead of ACS. Along the way, you'll see why Infobip is a natural next step for ACS customers: global reach, feature parity with what you rely on today, and a migration path built to minimize changes to your existing solution. You'll get a firsthand look at the tools available to help you prepare your solutions for a graceful migration away from ACS.
[eventID:27742]
In this webinar, you'll learn what Infobip and Microsoft have prepared to make your ACS migration smooth - from signup and onboarding with Infobip, to configuring your connection points back to your Azure solution, all the way to sending your first message through Infobip instead of ACS. Along the way, you'll see why Infobip is a natural next step for ACS customers: global reach, feature parity with what you rely on today, and a migration path built to minimize changes to your existing solution. You'll get a firsthand look at the tools available to help you prepare your solutions for a graceful migration away from ACS.
[eventID:27742]
![De datos sin procesar a datos preparados para IA: Spark en Fabric
La IA solo es útil si los datos que hay detrás también lo son.
En esta sesión, utilizaremos Spark en Fabric para tomar datos sin procesar, limpiarlos y transformarlos, y crear una base que una solución de IA o analytics pueda utilizar de verdad.
[eventID:27755] De datos sin procesar a datos preparados para IA: Spark en Fabric](https://i.ytimg.com/vi/_rNiSR-IUbg/mqdefault.jpg)
![From GitHub to GPU: Build an Open Source AI App with Copilot, Azure Linux, and AKS
What does it take to turn an AI idea into a real, scalable application without hiding the infrastructure behind a black box?
Well start with an open source GitHub repository and use GitHub Copilot to understand, adapt, and operate a complete AI workload. We deploy Ray and KubeRay on Azure Kubernetes Service (AKS), run LLM batch inference on an NVIDIA GPU, and explore how Azure Linux provides a lightweight, secure container host for the Kubernetes nodes under the hood.
Along the way, well connect the layers that make the application work using containers, Kubernetes resources, GPU scheduling, and open source operators.
Well also explore autoscaling, network isolation, authentication, metrics, dashboards, and cost-aware management. Leave with a reusable GitHub repo and a clear roadmap for building open source AI systems on Azure.
🔗 Learn more by exploring the resources: https://aka.ms/MicrosoftAzureOpenSourceBlog
📌 This event is part of a series. Learn more here: https://aka.ms/AIwithOpenSourceSeries/y
0:00 Introduction and housekeeping
0:50 Welcome and overview of the AI stack
3:14 Project repository and dashboard introduction
5:34 How GitHub Copilot was used to build the app
14:16 What is Ray and why use it on Kubernetes?
16:37 Intro to KubeRay and GPU scheduling
19:12 Building the platform with AKS and T4 GPUs
23:25 Anatomy of a GPU Ray cluster
26:22 Running GPU tests: Interactive vs. Declarative
27:54 Demo: VS Code dashboard and live inference
34:25 Understanding autoscaling layers
40:50 Metrics, observability, and dashboard breakdown
43:40 Network security and access policies
46:27 Authentication with cluster tokens
48:26 Gotchas and challenges from the POC
49:26 Cost awareness and billing tips
52:05 Summary of known issues and roadmap
55:12 Final roadmap to production
#MicrosoftReactor #LearnConnectBuild #OpenSource #Linux #Kubernetes #AIapps
[eventID:27664] From GitHub to GPU: Build an Open Source AI App with Copilot, Azure Linux, and AKS](https://i.ytimg.com/vi/_uybxmIaQiw/mqdefault.jpg)
![AI @Work: Community Builders in Action
ACFabric IQ sounds important, but what is it actually doing? We’ll break down how it uses business context to help AI understand your data, how Copilot fits in, and what grounded answers could look like in real work. Bring your questions.
📌 This event is part of a series. Learn more here:
https://aka.ms/DataDays/FabConRecap/y
https://aka.ms/DataDays/IntoCerts/y
https://aka.ms/AIatWork/UnderstandIt/y
https://aka.ms/AIatWork/BuildIt/y
https://aka.ms/AIatWork/ShowIt/y
[eventID:27725] AI @Work: Community Builders in Action](https://i.ytimg.com/vi/aDFvcZGlCSk/mqdefault.jpg)
![Copilot como tu primer asistente de IA: diseña un asistente para una tarea real
¿Qué pasaría si no tuvieras que explicarle a Copilot desde cero lo que necesitas cada vez que trabajas con él?
En esta sesión práctica aprenderemos a diseñar un asistente de IA especializado utilizando Microsoft Copilot. Partiremos de una necesidad concreta y construiremos paso a paso su propósito, contexto, instrucciones, tareas, límites y resultados esperados.
Trabajaremos con una estructura sencilla: Propósito → Rol → Contexto → Instrucciones → Tareas → Límites → Resultado esperado El objetivo no será construir un agente autónomo complejo, sino comprender y experimentar con los elementos fundamentales que necesitamos para convertir Copilot en un asistente diseñado para un propósito específico.
🔗 Aprende más accediendo a los recursos:
https://aka.ms/PromptAssitant
https://aka.ms/PromptGalleryLearn
📌 Este evento es parte de una serie, aprende más aquí: https://aka.ms/BuildWithCopilot/serie
#MicrosoftReactor #LearnConnectBuild #Copilot
[eventID:27608] Copilot como tu primer asistente de IA: diseña un asistente para una tarea real](https://i.ytimg.com/vi/arFmM5xQ_yI/mqdefault.jpg)
![Model Mondays - Spotlight On Fine-Tuned Models
Fine-tuning adapts a model to the language, examples, and behaviors of a specific task, helping developers improve quality, reduce prompt complexity, and right-size performance and cost. Join us as we explore the latest fine-tuning capabilities in Microsoft Foundry and see how customized models can strengthen specialized AI agent experiences.
0:00:03 - Introduction and Code of Conduct
0:00:54 - Speaker Introductions
0:03:40 - Overview of Microsoft AI (MAI) Models
0:06:29 - Five Classes of MAI Models: Image, Voice, Transcribe, Cyber, and Reasoning
0:09:15 - Real-world Applications of MAI Models
0:10:17 - Deep Dive into Reasoning and Coding Models
0:11:40 - Frontier Tuning Explained
0:12:53 - How Frontier Tuning Works: A Continuous Learning Loop
0:14:17 - Case Study: Land OLakes and Microsoft Internal Usage
0:15:08 - Step-by-Step Playbook for Frontier Tuning
0:17:24 - Demo: Shopping Bench Data Set
0:26:04 - Setting Up Hosted Agents in VS Code
0:27:59 - Demo: Running and Optimizing Agents in Foundry
0:33:36 - Demo: Reinforcement Learning for Web Tasks
0:41:46 - Q&A: RFT vs. SFT and Grader Calibration
0:44:04 - Q&A: Agent Optimizer Capabilities
0:46:40 - Q&A: Getting Started with Microsoft Foundry
0:49:37 - Weekly News Highlights: New MAI Models, Foundry IQ, Cohere, Anthropic, and GPT-6.1
- Explore the Resources: https://github.com/microsoft/model-mondays
- Read the Newsletter: https://aka.ms/model-mondays/newsletter
- Continue the conversation on the Discord: https://aka.ms/model-mondays/discord
[eventID:27542] Model Mondays - Spotlight On Fine-Tuned Models](https://i.ytimg.com/vi/b-82zc3LdMM/mqdefault.jpg)
![Episode 2: Generative AI – Creating, Innovating, and Accelerating with AI
Explore how Generative AI can create content, generate ideas, solve problems, and boost productivity. Participants will learn how AI is transforming creativity, collaboration, and innovation across industries.
🔗 Learn more by exploring the resources: https://aka.ms/aiconceptsfordevs
📌 This event is part of a series. Learn more here: https://aka.ms/aiskillsfortomorrow/y
0:00 Introduction and Welcome
3:00 Copilot Basics & Weather Example
13:00 Personalizing Copilot (Memories & Customization)
25:30 Using Copilot Agents (Idea Coach)
34:00 Using Copilot Agents (Researcher)
44:00 Daily Work Planning & Automation
52:00 Scheduling Prompts
54:00 Wrap-up & Future Episodes
#MicrosoftReactor #LearnConnectBuild #MicrosoftCopilot #Copilot #Agents #AIagents #GenAI
[eventID:27653] Episode 2: Generative AI – Creating, Innovating, and Accelerating with AI](https://i.ytimg.com/vi/bfBIwNhv6z4/mqdefault.jpg)
![Episode 1: AI Fundamentals – Understanding the Future of Work
Discover the fundamentals of Artificial Intelligence and explore how AI is transforming industries, careers, and everyday life. This introductory session demystifies core AI concepts, highlights real-world applications, and helps participants build a strong foundation for their AI journey.
🔗 Learn more by exploring the resources: https://aka.ms/aiconceptsfordevs
📌 This event is part of a series. Learn more here: https://aka.ms/aiskillsfortomorrow/y
0:00:06 Introduction and Housekeeping
0:00:59 Speaker Introduction
0:03:13 Session Overview and Goals
0:04:18 AI Fundamentals: Data and Core Pillars
0:06:15 Evolution: Rules to AI Models
0:09:05 Six Key AI Capabilities
0:12:15 Prompt Engineering Demo 1 (Scenario)
0:21:05 How AI Works (Backend Logic)
0:22:42 Sentiment Analysis Demo
0:29:25 Framework for AI Usage (Automate, Augment, Redesign)
0:32:45 Retrieval Augmented Generation (RAG) Demo
0:34:52 Responsible AI Principles
0:37:10 Practical Steps to Practice AI
0:39:08 Closing Remarks and Q&A
#MicrosoftReactor #LearnConnectBuild #MicrosoftCopilot #Copilot #Agents #AIagents #GenAI
[eventID:27652] Episode 1: AI Fundamentals – Understanding the Future of Work](https://i.ytimg.com/vi/dBKuN3qJrag/mqdefault.jpg)
![Capacidades avanzadas de MCP: Apps, Tasks y WebMCP
Ve más allá de las herramientas tradicionales de MCP y descubre cómo llevar el protocolo a nuevos escenarios. Explora MCP Apps, MCP Tasks y WebMCP para incorporar interfaces interactivas, gestionar operaciones de larga duración e integrar aplicaciones web con agentes de IA.
🔗 Aprende más accediendo a los recursos:
https://aka.ms/MCPrepo
https://aka.ms/IntroMCP
https://aka.ms/MCPtools
https://aka.ms/MCPagents
📌 Este evento es parte de una serie, aprende más aquí: https://aka.ms/mcpafondo
0:00 Introducción y bienvenida
1:53 ¿Qué son las extensiones MCP?
6:32 Explorando MCP Apps: Interfaces interactivas
12:12 Demo: MCP Apps en Fabio Robotics
24:42 Introducción a MCP Tasks: Procesos asíncronos
29:12 Implementación de MCP Tasks y persistencia
34:08 Demo: MCP Tasks en acción
40:03 Breve introducción a WebMCP
44:13 Seguridad en MCP: Riesgos y mejores prácticas
48:43 Análisis de amenazas (OWASP Top 10 para MCP)
59:44 Conclusión y recursos
#MicrosoftReactor #LearnConnectBuild #MCP #Agents
[eventID:27603] Capacidades avanzadas de MCP: Apps, Tasks y WebMCP](https://i.ytimg.com/vi/dDE_iLNNBIA/mqdefault.jpg)
![Skill up on Microsoft Copilot Studio
Discover what to learn first, how to get started building with Copilot Studio, and how Agent Academy can help you develop practical AI builder skills.
🔗 Learn more by exploring the resources: https://aka.ms/agent-academy
0:04 Introduction and Housekeeping
0:55 Speaker Introductions
1:23 Session Agenda
2:22 Why Focus on AI Skills?
6:51 What is Copilot Studio?
10:32 What’s New in Copilot Studio
12:02 Understanding the GitHub Copilot Harness
13:24 Enhancements to the Building Experience
14:33 Apps in Copilot Studio
15:14 Cost Visibility for Makers
15:51 Using Hooks for Deterministic Logic
16:37 Publishing Agents (MCP and A2A)
17:19 New Model Choices
17:42 How to Get Started with Skilling
19:38 Technical Deep Dive: Agents and Harnesses
24:31 Comparing Harness Scenarios
28:13 The Agentic Loop Explained
29:10 Core Building Blocks (Instructions, Knowledge, Tools)
39:24 Deep Dive into Skills
43:10 Memories and Personalization
43:47 The Sandbox Capabilities
44:41 Agent Academy Walkthrough
59:52 Q&A Session
#MicrosoftReactor #LearnConnectBuild #Copilot Studio #Agents
[eventID:27597] Skill up on Microsoft Copilot Studio](https://i.ytimg.com/vi/dN_birVmIHs/mqdefault.jpg)
![How to trust the code you ship
In this session, we’ll walk through a single pull request end to end and show how GitHub helps your team close the gap between code that’s written and software that’s trustworthy.
You’ll learn how GitHub supports stronger security and quality outcomes without adding more pressure to developers or reviewers.
Youll learn:
How to get a comprehensive view of the vulnerabilities in your codebase—including areas traditional scanning tools might miss.
How GitHub Code Security helps you turn findings into validated, review-ready pull requests, so you can shrink your backlog.
Tangible next steps on how you can automatically apply standards and make trust a consistent part of the development process.
- 0:00 Introduction
- 0:42 The growing security challenge
- 3:01 Overview of solutions
- 4:36 Audience poll: Security pain points
- 5:16 Governance and security configurations
- 11:42 Creating repository rule sets
- 18:34 Addressing AI-generated code and backlog
- 20:28 AI scanning for repositories and PRs
- 24:06 Demo: Pull request security and quality scanning
- 32:18 Security backlog management
- 34:28 Using Agentic Autofix
- 37:16 Mean time to remediation improvements
- 43:41 Live Q&A and community questions
- 53:53 Closing remarks
[eventID:27685] How to trust the code you ship](https://i.ytimg.com/vi/dZh4NA0ol60/mqdefault.jpg)
![De las ideas a la acción con Copilot: convierte una idea en un experimento
Generar ideas con inteligencia artificial es fácil. El verdadero reto es identificar cuáles tienen potencial y decidir qué hacer con ellas.
En esta sesión práctica utilizaremos Microsoft Copilot como compañero de innovación para transformar un desafío en ideas, analizarlas con criterios claros, priorizar oportunidades y convertir una idea seleccionada en un pequeño experimento.
Construiremos juntos un proceso sencillo: Desafío → Ideas → Criterios → Priorización → Hipótesis → Experimento
🔗 Aprende más accediendo a los recursos:
https://aka.ms/PromptAssitant
https://aka.ms/PromptGalleryLearn
📌 Este evento es parte de una serie, aprende más aquí: https://aka.ms/BuildWithCopilot/serie
#MicrosoftReactor #LearnConnectBuild #Copilot
[eventID:27610] De las ideas a la acción con Copilot: convierte una idea en un experimento](https://i.ytimg.com/vi/dvTeciJ5DNM/mqdefault.jpg)