Jose Manuel Martinez Lago
Power Platform Solution Architect @ BE-Terna
Actions
José Martínez Lago, a Spanish national residing in Austria, serves as a Power Platform Solution Architect at BE-terna. In this role, he enhances business solutions using Microsoft Power Platform and Dynamics 365, focusing on user-centric design and seamless integration. Additionally, José is a Microsoft Certified Trainer (MCT) specializing in Power Platform.
Oops... I prompt it again: CodeApps con Copilot y el drama de a ver quién mantiene esto
Admitámoslo: programar con GitHub Copilot engancha. Con Vibe CodeApps, generative pages y GitHub Copilot, la IA empieza a escupir código y, de repente, funciona. La barrera técnica se ha desplomado y la creación de componentes avanzados se ha democratizado. El problema es que los prototipos en producción los carga el diablo si nadie sabe qué hay bajo el capó. ¿Cómo escalamos esto a nivel empresarial sin morir en el intento?
En esta sesión puramente práctica abordaremos el ciclo de vida completo de las Vibe CodeApps y Generative Pages utilizando IA generativa de extremo a extremo. Mediante un enfoque dinámico que combina explicaciones estratégicas, demostraciones reales y un bloque Hands-on, pasaremos por las siguientes fases críticas:
Inception & Planning: Cómo usar IA y prompts avanzados para rellenar tareas de DevOps, definir el alcance y organizar el backlog del proyecto.
Implementation: Desarrollo ágil de CodeApps asistido por GitHub Copilot, entendiendo qué genera la IA y cómo evitar la deuda técnica inmediata.
Deployment and ALM: Estrategias y automatización del ciclo de vida de la aplicación para empaquetar, desplegar y asegurar que el código generado sea mantenible, seguro y gobernable antes de pisar los entornos de producción.
Understand What Users Actually Do, With Clarity Heatmaps in Dynamics 365
Have you ever published a model-driven app and wished you could see what users actually do—where they click, hesitate, or give up?
Clarity delivers heatmaps and session replays for model driven apps, canvas apps, and even power pages! This tool may not be available out of the box, but in this session we will help you set it up so you can capture real user behaviour without redesigning your forms.
We’ll cover where Clarity adds the most value in Dynamics scenarios, what to track (and what not to), and how to stay compliant with masking and consent. Then we’ll walk through a quick, repeatable install you can apply to any environment.
You’ll leave knowing when to use Clarity, how to roll it out safely, and how to turn those insights into backlog-ready UX improvements, all backed by a short live demo you can replicate in minutes!
LLM Showdown: Local vs. Cloud + MCP
Are you ready to explore the rapidly evolving world of Large Language Models (LLMs)? This session will pit two powerful approaches against each other – running LLMs locally using LM Studio versus leveraging cloud-based services. We’ll dive deep into the pros and cons of each, examining factors like cost, performance, control, and security.
What you'll learn:
LM Studio Deep Dive: Discover how to effectively utilize LM Studio for running LLMs directly on your hardware – exploring its features, model selection, and optimization techniques.
Cloud LLM Landscape: We’ll compare popular cloud-based LLM providers (like Azure OpenAI Service) discussing their pricing models, API access, and scalability options.
MCP Integration - The Core Focus: This session will demonstrate how to seamlessly connect both local LM Studio LLMs and cloud-based LLMs to your Microsoft Cognitive Platform (MCP) servers. We’ll cover:
Connecting via APIs – showcasing practical examples.
Utilizing community-based tools and frameworks designed for MCP integration with LLMs (e.g., LangChain, LlamaIndex).
Exploring the potential of Microsoft MCP Servers for hosting and managing your LLM deployments.
Cost Analysis: A brief overview of the cost implications of each approach – considering hardware investment vs. cloud usage fees.
Harder, Better, Faster, Stronger: Enterprise Interfaces using Azure
Has your data ever felt the need for speed, but you didn't have the right tools to provide it? One of the biggest mistakes in cloud architecture is choosing technology before understanding the integration pattern. Should the solution be synchronous or asynchronous? Real-time or near real-time? This session offers a practical framework for answering those questions by focusing on the trade-offs between Azure integration tools, and the specific situations where each one makes sense.
Taking Logic Apps and Azure Functions as a base, we will demonstrate how to harden your connectivity by layering API Management, a dedicated Security Layer, and Log Analytics to ensure your solution is reliable, supportable, and fit for production. Then, using Dataverse as a concrete example, we will explore principles that provide a universal blueprint to scale the solution to anything from Copilot Studio to Sharepoint. Attendees will leave with a clearer way to evaluate requirements and design Power Platform integrations that are not just effective, but maintainable and easy to operate.
¿Cómo extraer datos de uso de tu Model Driven App? ¡Microsoft Clarity al rescate!
¿Te gustaría comprender cómo interactúan realmente los usuarios con tu Model Driven App? Aunque esta herramienta no ofrece una integración nativa con Model Driven Apps, existen soluciones creativas para obtener información valiosa sobre el comportamiento del usuario. En esta sesión, exploraremos cómo aprovechar JavaScript para integrar Clarity en tus formularios, permitiéndote visualizar mapas de calor, grabaciones de sesiones y detectar áreas problemáticas. Además, discutiremos cómo complementar estos datos con Copilot para obtener una visión más completa del rendimiento y uso de tus aplicaciones. ¡Descubre cómo transformar datos en decisiones informadas y mejorar la experiencia del usuario!
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top