Rob Kerr's Blog
BigQuery → Fabric Mirroring Made Easy (Automation Script Included) 6:26

BigQuery → Fabric Mirroring Made Easy (Automation Script Included)

Microsoft Fabric

This video walks through the end-to-end process of setting up Microsoft Fabric database mirroring from Google BigQuery. Viewers will learn how to prepare a BigQuery environment using an automated script for service account creation and IAM permissions, and then how to configure the Fabric connection, set up a mirrored database, and initiate replication. The presenter demonstrates how to monitor sync status at the table level and explains the underlying mechanics, including near real-time incremental sync and the conversion of data to Delta tables in OneLake for immediate use with SQL, Power BI, and AI scenarios.

Jun 24, 2026140 views
Query BigQuery Iceberg Tables from Microsoft Fabric (Shortcuts + Automation) 5:20

Query BigQuery Iceberg Tables from Microsoft Fabric (Shortcuts + Automation)

Microsoft Fabric

This video walks through how to use Microsoft Fabric shortcuts to access Google BigQuery Iceberg tables without copying or moving data. It demonstrates a hands-on approach, including an automation script that prepares the Google Cloud side for seamless integration. Viewers will learn how Fabric shortcuts work, how to configure connections, and the differences between using shortcuts versus mirroring for external data access.

Jun 21, 2026123 views
Building NFL Analytics with Fabric, Power BI, and AI Agents 33:16

Building NFL Analytics with Fabric, Power BI, and AI Agents

Azure AI FoundryMicrosoft Fabric

This video walks through building an end-to-end NFL analytics reference architecture in Microsoft Fabric. Viewers will learn how to download NFLVerse data, set up a Fabric workspace and Lakehouse, and transform raw files into curated Gold star schemas. The workshop also demonstrates two Fabric Data Agent patterns—one using SQL over Gold tables and another using DAX over a Power BI semantic model—and concludes with testing the agents in Microsoft 365 Copilot, running evaluations, and iterating on the development loop.

May 25, 2026320 views
From On Prem SQL to Natural Language Queries with Fabric Mirroring & Data Agents 7:29

From On Prem SQL to Natural Language Queries with Fabric Mirroring & Data Agents

Microsoft Fabric

This video walks through how to expose on-premises SQL Server data through a natural language interface using Microsoft Fabric Mirroring and OneLake. Viewers will see how to set up Change Data Capture and the On-Premises Data Gateway to mirror selected tables into OneLake as Delta tables, and then learn how to build a Fabric Data Agent that allows users to query the data using plain English without needing to write SQL.

Apr 20, 2026256 views
Microsoft Fabric Private Link: Secure a Workspace in an Azure VNet 17:22

Microsoft Fabric Private Link: Secure a Workspace in an Azure VNet

Microsoft Fabric

This video walks through configuring a Microsoft Fabric workspace with a Workspace-scoped Private Link to restrict access to an Azure Virtual Network. It provides a complete end-to-end walkthrough, demonstrating how to build the solution from scratch, including setting up the VNet, subnets, private endpoint, and a jump box VM for testing. Viewers will learn how to implement a practical architecture pattern that combines strong network isolation with flexible reporting access, enabling restricted Lakehouse data to be securely queried by open reporting workspaces via a Virtual Network Data Gateway.

Mar 16, 20261,078 views
Using Python to Build a Microsoft Fabric Ontology 4:31

Using Python to Build a Microsoft Fabric Ontology

Microsoft Fabric

This video walks through building a Microsoft Fabric ontology using Python code within a notebook. It demonstrates how to define entities, keys, and relationships programmatically, use PySpark to read lakehouse schemas, and construct REST API payloads to automate deployment and validation.

Mar 08, 2026758 views
Microsoft Fabric Real-Time: Streaming IoT Truck Data into an Ontology 3:09

Microsoft Fabric Real-Time: Streaming IoT Truck Data into an Ontology

This video walks through a real-time streaming pipeline in Microsoft Fabric designed to process telemetry data from trucking equipment. Viewers will see how to ingest IoT events using Event Streams, filter for mechanical faults as they occur, and save critical operational events to a Lakehouse table. The demonstration also covers triggering automated workflows via Fabric Activator and visualizing live data moving through an ontology-driven model.

Mar 06, 2026225 views
Fabric Data Agent + Foundry Agent: Building a Multi-Source AI Assistant 1:55

Fabric Data Agent + Foundry Agent: Building a Multi-Source AI Assistant

Microsoft Fabric

This video walks through the integration of a Microsoft Fabric Data Agent with a Microsoft Foundry Agent to create a multi-source AI assistant. It demonstrates how to query structured enterprise data through a Fabric Ontology, calculate operational metrics like trip load values, and answer questions from unstructured company documentation using an Azure AI Search vector index. The presentation concludes by showing how Foundry orchestrates these diverse knowledge sources and how the resulting agent can be embedded into custom applications to support natural language queries.

Mar 06, 2026434 views
Can Microsoft Fabric Power Custom Apps? (Yes — Here’s How) 4:23

Can Microsoft Fabric Power Custom Apps? (Yes — Here’s How)

Microsoft Fabric

This video walks through building a secure, production-style web application that integrates Microsoft Fabric with custom app development. It demonstrates how to combine a React and Next.js frontend with a FastAPI Python backend to deliver map-based visualizations and generative AI experiences, utilizing Fabric Graph API, Fabric Ontology, and Microsoft Foundry Agents without data duplication.

Mar 02, 2026426 views
Designing and Building an Ontology in Microsoft Fabric (uncut) 41:08

Designing and Building an Ontology in Microsoft Fabric (uncut)

Microsoft Fabric

This video walks through the fundamentals of data modeling, explaining what an ontology is and how it combines data, logic, and action. It provides a hands-on guide to designing entity types and relationships using best practices, then demonstrates the step-by-step process of building and implementing an ontology within Microsoft Fabric.

Feb 17, 20268,862 views
Microsoft Fabric Graphs Explained: Graph Models, GQL, and Real Analytics Use Cases 9:04

Microsoft Fabric Graphs Explained: Graph Models, GQL, and Real Analytics Use Cases

Microsoft Fabric

This video provides a clear, beginner-friendly introduction to graph data analysis within Microsoft Fabric. It walks through the fundamentals of graph models, explaining how nodes, edges, and relationships unlock new analytical questions, and demonstrates how to build and query a simple graph using customer and order data. Viewers will learn about Fabric’s current graph capabilities, including GQL, and see how to access these features directly from a Fabric Notebook.

Feb 10, 20261,628 views
Microsoft Foundry Local -- First Look and Walk Through 10:58

Microsoft Foundry Local -- First Look and Walk Through

This video walks through Microsoft Foundry Local, a local inference engine introduced at Microsoft Ignite 2025. It provides a hands-on look at how the runtime, API layer, and SDK work together to enable running large language models directly on Windows, macOS, or Windows Server 2025. Viewers will follow a step-by-step installation on Windows 11 and see live demonstrations of testing inference performance on an NVIDIA RTX 4090, as well as guidance on building local applications using VS Code and an OpenAI-compatible SDK.

Dec 07, 20251,796 views
Building Custom MCP Tools for Azure AI Foundry Agents (with Cosmos DB GraphRAG) 20:04

Building Custom MCP Tools for Azure AI Foundry Agents (with Cosmos DB GraphRAG)

This video walks through building a complete flow for using custom MCP tools with an Azure AI Foundry Agent. It demonstrates how to design and implement a Python-based MCP server that exposes GraphRAG tools over HTTP, deploy it as an Azure Function, and wire it into an Azure AI Foundry Agent to query a Cosmos DB Gremlin graph. Viewers will learn how to wrap custom Python logic as MCP tools, host them in a lightweight cloud service, and enable agents to call into a GraphRAG backend via standard MCP protocols rather than bespoke REST endpoints.

Dec 01, 20254,359 views
Building a GraphRAG App: Concepts, Code, and Demo 18:10

Building a GraphRAG App: Concepts, Code, and Demo

NVIDIA DGX Spark

This video provides a clear, accessible walkthrough of GraphRAG technology, explaining what graph databases are and why they are used to model real-world relationships. It walks through the mechanics of how GraphRAG improves LLM retrieval through structure and context, compares it to traditional Vector RAG, and concludes with a full-code demonstration of a functional application.

Nov 27, 20252,810 views
Vectorizing Images with LLMs [Full Stack Deep Dive] 13:04

Vectorizing Images with LLMs [Full Stack Deep Dive]

NVIDIA DGX Spark

This video walks through how to build a complete end-to-end image vectorization and similarity search system using modern AI tools, multimodal LLMs, and GPU-powered infrastructure. Viewers will learn how to represent images as vectors, generate embeddings using multimodal models, and store them efficiently using ChromaDB, while observing a live demonstration of the final architecture in action.

Nov 20, 20251,455 views
From Text to SQL: Fine-Tuning Phi-4-Mini-Instruct with LoRA and PyTorch (Deep Dive) 37:19

From Text to SQL: Fine-Tuning Phi-4-Mini-Instruct with LoRA and PyTorch (Deep Dive)

NVIDIA DGX Spark

This video walks through the full workflow of fine-tuning the Phi-4-Mini-Instruct model using LoRA and PyTorch to improve natural-language-to-SQL accuracy. Viewers will learn how to prepare custom training data, code a PyTorch fine-tuning script, and run jobs within a Docker container. The demonstration also covers merging adapters, quantizing the model for edge deployment, and testing the final results using Llama.cpp and LM Studio.

Nov 09, 20251,017 views
Build Smarter AI Systems: 3 Agent Patterns You Can Implement Today 9:39

Build Smarter AI Systems: 3 Agent Patterns You Can Implement Today

This video walks through three distinct Agentic AI patterns—Chat Agents, Workflow Agents, and Autonomous Agents—and demonstrates how to implement each one. Viewers will learn the fundamental differences between AI agents and chatbots, how large language models and tools interact to enable reasoning, and the specific use cases for each pattern to solve real-world problems.

Nov 05, 2025144 views
NVIDIA DGX Spark — the data-center-grade AI lab that fits on your desk, but is it right for you? 5:47

NVIDIA DGX Spark — the data-center-grade AI lab that fits on your desk, but is it right for you?

NVIDIA DGX Spark

This video walks through the NVIDIA DGX Spark, a compact AI system designed to bring data-center-grade capabilities to a desktop environment. It provides a hands-on overview of the hardware specifications, evaluates the value proposition for different users, and offers a summary of the presenter's real-world experience testing the system for building, testing, and deploying AI prototypes.

Oct 27, 202513,972 views
How to configure Entra ID SSO for Snowflake 18:06

How to configure Entra ID SSO for Snowflake

Azure Administration

This video walks through the step-by-step configuration of a Microsoft Azure Entra Application Registration to enable users to access a Snowflake database using their Entra ID credentials. It demonstrates the process within a React single-page web application, illustrating how the techniques apply to virtually any application capable of integrating with Entra via MSAL or REST connections.

Jul 09, 20251,430 views
Query Fabric Semantic Models from React Web Applications 7:27

Query Fabric Semantic Models from React Web Applications

Microsoft Fabric

This video walks through the process of integrating a Microsoft Fabric Semantic Model query into a custom web application built with React and TypeScript. It covers the necessary configuration steps, including setting up API permissions and app registrations, and demonstrates how to fetch access tokens and implement REST API calls to query and display semantic model results within a React interface.

Mar 01, 2025713 views
Tutorial: Implementing Entra MSAL with a React Web Application 31:03

Tutorial: Implementing Entra MSAL with a React Web Application

Azure Administration

This video walks through the process of configuring an Entra App Registration and implementing authentication in a React web application from scratch. Viewers will learn how to set up an app registration, configure redirect URLs and user restrictions, and implement the login, logout, and token refresh flows using the MSAL library.

Feb 27, 20255,130 views
Deploy an Azure Web App via an Azure DevOps CD Pipeline with a custom App Registration 8:38

Deploy an Azure Web App via an Azure DevOps CD Pipeline with a custom App Registration

Azure Administration

This video walks through the process of deploying an Azure Web App using a custom Azure DevOps continuous deployment pipeline. It provides a step-by-step guide on configuring the necessary Entra ID app registration, creating a client secret, and setting up deployment permissions before demonstrating how to build and run the pipeline to test a Python application.

Jan 27, 20251,608 views
Windows Fine Tuning Combined Streams 18:16

Windows Fine Tuning Combined Streams

This video walks through the process of setting up a Windows-based development workstation for fine-tuning large language models. It provides a step-by-step guide covering the installation of Windows Subsystem for Linux, essential tools like g++, Miniconda, and VS Code, and the configuration of Unsloth components to enable efficient model training.

May 10, 20241,486 views
Unleash Your Model's Potential: Guide to Deploying a Fabric ML Model on Azure ML Inference Endpoint 6:23

Unleash Your Model's Potential: Guide to Deploying a Fabric ML Model on Azure ML Inference Endpoint

Azure AI FoundryMicrosoft Fabric

This video walks through the steps to export a model developed in Microsoft Fabric, import it into an Azure Machine Learning workspace, and deploy it to a real-time inference endpoint. Viewers will learn how to move a model from a Fabric Data Science workload to Azure ML, including the process of creating and testing the endpoint both within the Azure ML interface and via Postman.

Feb 27, 20241,096 views
How to incorporate AI-Driven Sentiment Analysis models in a Fabric Data Lake Solution 7:09

How to incorporate AI-Driven Sentiment Analysis models in a Fabric Data Lake Solution

This video walks through the process of integrating Azure AI sentiment analysis models into a Fabric Data Lake solution. Viewers will learn how to implement machine learning models within a Jupyter notebook environment to extract sentiment features from unstructured data, enriching Data Lake tables with quantitative insights for data analysts.

Jan 08, 2024368 views
Create Gemini Chatbot 9:55

Create Gemini Chatbot

This video walks through building a functional Gemini Chatbot using the Streamlit framework. It covers the setup process, including installing dependencies and initializing the Gemini SDK, before demonstrating how to create a history object and user interface prompts to run the bot.

Dec 16, 20234,871 views
Creating a Custom Generative AI Microsoft Teams Copilot using Copilot Studio 18:58

Creating a Custom Generative AI Microsoft Teams Copilot using Copilot Studio

This video walks through the process of creating a custom Microsoft Teams copilot using Microsoft Copilot Studio. It demonstrates how to build the assistant from scratch, covering the setup of generative answers using web and PDF documentation, as well as the creation of custom topics with branching logic. The walkthrough concludes with publishing the copilot and deploying it to a Teams channel for testing.

Dec 04, 20238,557 views
Build a Generative AI Chatbot Using a Vector Database for Custom Data (RAG) 14:39

Build a Generative AI Chatbot Using a Vector Database for Custom Data (RAG)

This video walks through building a generative AI chatbot that queries OpenAI's Large Language Model using a vector database. It demonstrates how to leverage Retrieval Augmented Generation (RAG) to introduce semantic understanding of custom data, utilizing OpenAI GPT models, Pinecone, and Python Notebooks to process and query source information.

Nov 05, 20233,625 views
Bring your own data to Azure OpenAI Service Large Language Models 6:52

Bring your own data to Azure OpenAI Service Large Language Models

This video walks through the implementation of Retrieval Augmented Generation (RAG) using Microsoft Azure OpenAI Service. Viewers will learn how to upload private data, configure OpenAI deployments, and create a RAG index to enable Generative AI queries that incorporate custom datasets.

Oct 31, 2023907 views
Azure ML Deploy Inference Endpoint 19:11

Azure ML Deploy Inference Endpoint

Azure AI Foundry

This video walks through the process of deploying a machine learning model to a production-ready REST endpoint. It provides a step-by-step tutorial on creating the necessary Azure resources, training a model using a notebook environment, and configuring it for real-time inference. Viewers will learn how to register the trained model and test the deployed endpoint via HTTP requests.

Oct 24, 20233,186 views
Exporting Azure AI Vision Models to run on iOS Devices 13:48

Exporting Azure AI Vision Models to run on iOS Devices

Azure AI Foundry

This video walks through the process of exporting AI models from Azure AI Vision and running them locally on iOS devices. It provides a step-by-step guide covering the Azure Portal, testing the model, exporting to a CoreML package, and integrating the code into an Xcode project to run the application on an iPhone.

Oct 08, 2023323 views
Using AI Vision to Detect Packaging Defects 10:50

Using AI Vision to Detect Packaging Defects

Azure AI Foundry

This video walks through the process of creating an Azure AI Custom Vision object detection model to identify packaging defects. It provides a step-by-step guide, from setting up Azure resources and importing images to tagging, training, and testing the model.

Oct 05, 2023942 views