Samantha Cruz
Data Engineering Team Lead @isolutions AG
Barcelona, Spain
Actions
I’m a Data Engineering Team Lead at isolutions Barcelona. I lead a team building data platforms for analytics, reporting, and AI using Azure, Microsoft Fabric, Databricks, and Power BI. My role combines technical architecture, project delivery, team leadership, and direct collaboration with clients. I also speak at industry and community events, sharing lessons from real projects, including what worked, what failed, and what we changed.
Area of Expertise
Topics
De Fabric a Azure DevOps: agentes que leen, escriben y documentan por ti
En el día a día de un equipo de datos, gran parte del tiempo no se va en escribir código, sino en todo lo que rodea el código: actualizar el board de Azure DevOps, documentar en la wiki, dejar trazabilidad de lo que se hizo y por qué. Son tareas necesarias, pero repetitivas, y normalmente las dejamos para "luego" y luego no llega.
En esta sesión cuento cómo hemos integrado agentes de IA (con GitHub Copilot y Agent Mode) en nuestro flujo de trabajo real sobre Microsoft Fabric, para que el propio agente pueda:
Leer y actualizar work items en el board de Azure DevOps
Escribir documentación técnica directamente en la wiki de DevOps
Mantener sincronizado el trabajo en Fabric con su seguimiento en DevOps, sin que tengamos que hacerlo manualmente cada vez
Retos reales que nos encontramos al conectar estas piezas (permisos, contexto, cuándo confiar en el agente y cuándo no), y las decisiones que tomamos para que esto funcione sin añadir complejidad innecesaria.
Si gestionas datos en Fabric y sientes que la documentación y el tracking siempre quedan atrás, esta sesión te da un punto de partida concreto para automatizarlo con agentes, sin perder control sobre lo importante.
Agentes de IA para meros mortales
¿Cuándo tiene realmente sentido construir un agente de IA y cuándo solo estás añadiendo una complejidad innecesaria?
La IA ya forma parte de cómo trabajamos, pero la mayoría de las personas todavía la utilizan como un chat o una herramienta de autocompletado. El siguiente paso es pasar de pedir ayuda a dejar que la IA realice partes del trabajo por ti. Ahí es donde entran los agentes.
En esta sesión exploramos cómo construir y utilizar agentes de IA de forma práctica usando herramientas como GitHub Copilot, Agent Mode, MCP y habilidades reutilizables. El enfoque no está en arquitecturas complejas, sino en lo que resulta útil en el trabajo del día a día.
Te guiaremos a través de qué es un agente de IA y por qué tiene sentido usar uno en lugar de solo prompts o automatizaciones simples. Verás cómo los agentes pueden razonar, planificar y ejecutar tareas en múltiples pasos, y cómo aplicar patrones que ya funcionan en producción.
También analizamos cómo encaja esto en tu flujo de trabajo actual, especialmente en entornos como VS Code, y cómo los agentes pueden conectarse a sistemas y datos reales. Ejemplos prácticos mostrarán cómo automatizar documentación, interactuar con plataformas de datos como Microsoft Fabric y crear flujos de trabajo reutilizables que ahorran tiempo.
La sesión también cubre errores comunes, incluyendo complicar en exceso las soluciones con demasiados agentes, y cómo mantener las cosas simples, controladas y mantenibles.
El objetivo es darte una comprensión clara y práctica para que puedas empezar a usar agentes en tu propio trabajo sin sobreingeniería.
Sin hype. Datos reales. Criterios claros.
How Much Does a Prompt Cost?
AI often feels simple: you write a prompt, you get an answer. But behind that interaction, multiple systems are running, consuming resources, and generating real cost.
In this session, we go under the hood of AI agents and Copilot experiences to understand what actually happens when a prompt is executed. We break down how tokens, queries, pipelines, and Microsoft Fabric capacity contribute to the final cost, and how these costs accumulate at scale.
Through practical scenarios, we compare different approaches, from simple prompts to more complex data-driven questions, showing how design and architecture decisions directly impact cost and performance. We will also discuss common misconceptions and where teams tend to underestimate expenses.
By the end of the session, you will have a clear mental model of how AI costs are generated, how to translate usage into real money, and how to design more efficient and predictable AI solutions in production.
AI-Powered Multilingual Reports with Azure AI and Semantic Link Labs
Creating Power BI reports for multilingual audiences can become complex and repetitive. This session presents a scalable and automated approach to building dynamic, multi-language reports using Azure AI Translator and Semantic Link Labs.
Semantic Link Labs will be used to translate metadata elements such as table names, column labels, and measure descriptions directly within the Power BI semantic model. Meanwhile, Azure AI Translator will process and translate the content inside tables, for example, values in dimension tables that need to appear in multiple languages in slicers or visuals.
Topics covered will include:
- Translating model metadata using Semantic Link Labs
- Translating table content dynamically with Azure AI Translator
- Enabling language switching based on user browser language
- Advantages of using multi language reports
- Deployment and automation strategies for multilingual scenarios
This session is ideal for data teams building Power BI reports for international users, or organizations looking to streamline multilingual reporting without duplicating content.
Designing and Operationalizing Data Agents within Microsoft Fabric
This session explains how Data Agents function inside Microsoft Fabric and how they support data connectivity, orchestration, and automated actions. You will see the main building blocks that enable these agents and walk through a practical setup of an agent in Fabric. We will also make a detailed comparison between Fabric Data Agents and other comparable tools, outlining similarities, differences, and where each option is most suitable.
AI in the Data Platform: Genie vs Fabric Data Agents in Real Architectures
Most data teams already run a Lakehouse. The harder problem now is deciding how far AI should be embedded into the data platform itself, and what that means for governance, ownership, and operational complexity. Two tools are gaining attention in this subject: Databricks Genie and Microsoft Fabric Data Agents.
This session looks at Genie and Fabric Data Agents as platform-level design choices. Both promise natural language access and AI-driven interactions with data, but they rely on very different assumptions about metadata, semantic enforcement, and control boundaries. Those assumptions matter once teams operate multiple domains, shared Lakehouses, and strict governance requirements.
Using real architectural patterns, we examine how each approach fits into a production data platform built on Lakehouse principles. We discuss where AI logic lives, how trust in results is established, what breaks at scale, and how these tools interact with lineage, security models, and DataOps workflows. The comparison focuses on consequences, not capabilities.
The goal is to help data engineers and platform architects make informed decisions about embedding AI into the core of their data platform, understanding the trade-offs before these tools become foundational dependencies.
Integrando Dataverse y Microsoft Fabric
Esta sesión se centra en cómo Microsoft Dataverse se integra con Microsoft Fabric y qué se puede hacer con esta integración.
Se explicará cómo los datos de Dataverse se exponen en Fabric, cómo pueden consultarse y combinarse con otras fuentes de datos, y cómo esto habilita escenarios de analítica y reporting.
La sesión incluye demostraciones prácticas de casos de uso y flujos de trabajo habituales.
Gobernanza de Datos en Fabric: Lineage de Datos con Purview
En este proyecto se implementó una solución para registrar la trazabilidad de datos (data lineage) en Microsoft Purview desde notebooks de Spark en Microsoft Fabric. Automatizamos el registro de activos y procesos que ocurren dentro del Lakehouse, para que se pueda visualizar cómo los datos fluyen entre tablas y transformaciones.
Se utilizó la librería PyApacheAtlas, que permite interactuar con el catálogo de datos de Purview mediante el API de Atlas. A través de una función desarrollada en Python, se crean entidades que representan las tablas (por ejemplo, las capas Bronze, Silver y Gold), y también los procesos que conectan estas tablas, como ingestiones, transformaciones o fusiones de datos. Cada elemento queda registrado con su nombre, ubicación y descripción, y puede visualizarse luego en la vista de "lineage" de Purview.
Toda la autenticación se maneja mediante un Service Principal configurado en Key Vault, permitiendo que la solución funcione de forma segura dentro del entorno de Fabric.
Esta solución mejora la visibilidad, facilita auditorías y fortalece la gobernanza de los datos.
A modo de cierre, se presetarán posibles líneas de evolución para esta solución, como la integración de OpenLineage para capturar automáticamente la trazabilidad durante la ejecución de los notebooks.
AI-Powered Multilingual Reports with Azure AI and Semantic Link Labs
Creating Power BI reports for multilingual audiences can become complex and repetitive. This session presents a scalable and automated approach to building dynamic, multi-language reports using Azure AI Translator and Semantic Link Labs.
Semantic Link Labs will be used to translate metadata elements such as table names, column labels, and measure descriptions directly within the Power BI semantic model. Meanwhile, Azure AI Translator will process and translate the content inside tables, for example, values in dimension tables that need to appear in multiple languages in slicers or visuals.
Topics covered will include:
- Translating model metadata using Semantic Link Labs
- Translating table content dynamically with Azure AI Translator
- Enabling language switching based on browser locale or URL parameters
- Avoiding duplicated reports and reducing maintenance effort
- Deployment and automation strategies for multilingual scenarios
This session is ideal for data teams building Power BI reports for international users, or organizations looking to streamline multilingual reporting without duplicating content.
Deploy Your Microsoft Fabric Architecture with CI/CD in 10 Minutes
This session explores how to deploy a complete Microsoft Fabric architecture using Terraform and Azure DevOps.
The focus is on applying infrastructure as code (IaC) principles to data architecture, making every environment setup consistent, repeatable, scalable, and fast.
We start by introducing the structure of a Microsoft Fabric deployment, including capacities, workspaces, and associated resources. Then we look at how to model this architecture using Terraform, leveraging the Fabric provider to define Fabric-specific resources.
The Terraform code is modular and designed to be reused across different environments. By defining variables for project names, environment tags, and other settings, the same code can be used to deploy dev, test, and prod setups with minimal changes.
We configure Azure DevOps to run the Terraform code through multi-stage pipelines. These infra-structure pipelines handle validation, plan, apply, and include optional manual approval gates for production stages.
The full CI/CD flow ensures that deployments are automated, controlled, and aligned with best practices for managing infrastructure in enterprise environments.
A live demonstration shows the end-to-end process: cloning the code, updating variables, triggering the pipeline, and deploying the Microsoft Fabric components in Azure.
This approach allows data teams to treat infrastructure as code, integrate it into version control, and reduce the time and risk involved in setting up Azure and Fabric objects manually.
Queensland Power BI and Fabric User Group User group Sessionize Event
Microsoft Speakers Hub en Español | Jul. - Sept. 2026 Sessionize Event
AICD - Tunis 2026 Sessionize Event
AgentCamp Barcelona 2026 Sessionize Event
Global Power Platform Bootcamp Barcelona 2026 Sessionize Event
Power BI & Fabric Barcelona (Powered By Xarxa Telecos) User group Sessionize Event
How Do I fIT in? Women’s Journeys in Tech: from First Steps to Leadership.
This event was a special edition of the Isolutions Tech Meetup series focused on open and practical discussions about building a career in technology as a woman. It brought together women from different backgrounds, roles, and seniority levels to share real experiences, key career decisions, and lessons learned, creating a supportive space for knowledge exchange, inspiration, and professional networking. I participated as a panelist, sharing my own career journey in tech, including technical growth, challenges, and decision making along the way. During the panel discussion and the audience Q and A, I contributed practical insights and concrete examples, and I also engaged with participants during the networking moments, exchanging perspectives with women at different stages of their careers.
Bizz Summit ES 2025 Sessionize Event
Power BI & Fabric Barcelona (Powered By isolutions) User group Sessionize Event
Power BI & Fabric Summer Fest Barcelona Sessionize Event
Global Azure Barcelona Sessionize Event
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