-
Continue reading →: Your System Prompt Is Code. Start Treating It Like ItPrompt versioning is the least glamorous part of building agents in Microsoft Foundry, and the first thing you will wish you had done properly. Here is a situation I have seen more than once. An agent has been live for a few weeks and is behaving well. Someone in the…
-
Continue reading →: Debugging the Black Box: Why Tracing Is the First Thing You Should Turn On in Microsoft FoundryIf you’ve built more than one agent, you’ve had this moment. The agent gives a wrong answer, or a weird one, or just hangs for eight seconds before responding, and you have absolutely no idea why. Was it the retrieval step? A bad tool call? The model looping on itself?…
-
Continue reading →: Getting Started with Microsoft Foundry: How to Deploy Your First AI ModelIf you’re looking to get started with Microsoft Foundry (Microsoft’s unified platform for building and deploying AI models on Azure) this guide covers everything you need. In under 15 minutes, you can go from a blank Azure subscription to a running model deployment you can call from code. What Is…
-
Continue reading →: Your AI Shouldn’t Have to Phone Home (Intro to Foundry Local)Last week at MVP Summit I had the pleasure of joining a roundtable with the Microsoft Foundry Local team. It was a genuinely productive session, the kind that leaves you scribbling notes on the when you get home, and it’s sparked a deeper interest in what Foundry Local can actually…
-
Continue reading →: I’m BackAfter a year away, I’m returning to this blog. There’s no complicated way to say it. The past twelve months involved serious family illness and, ultimately, bereavement. Beyond the emotional toll, much of my time and energy went into caring responsibilities and being present for other family members who were…
-
Continue reading →: AI Ethics: A Growing PriorityAt several conferences I’ve attended recently, one topic keeps coming up with increasing urgency: AI ethics. Whether in panels, workshops, or casual conversations between sessions, business leaders, technologists, and strategists are all asking similar questions: How do we build responsible AI? What guardrails are needed? How can we ensure fairness,…
-
Continue reading →: The Rise of Small but Mighty: Why Tiny AI Models Are the Next Big ThingIn the race to build ever-larger AI models, a quiet revolution is gaining momentum—one that values efficiency over size, speed over scale, and privacy over power. While giants like GPT-4 and Claude 3 Opus continue to dominate headlines, a new generation of small yet powerful AI models is changing the…
-
Continue reading →: Meet MCP – The “USB-C for AI” ConnectionsImagine if all your gadgets—phone, laptop, camera—used completely different plugs. Frustrating, right? That’s how older AI systems felt, struggling to connect seamlessly with different apps and data sources. MCP, short for Model Context Protocol, changes that. It acts like a universal connection port for AI, letting multiple tools and systems…
-
Continue reading →: In Copilot we trust?I was recently at the Microsoft most valuable professional (MVP) summit in Redmond Seattle and the term automation bias was introduced to me during a talk and it was a concept that really caught my interest. So I decided to explore it in definition, it’s impacts and how we can…
Hello,
I’m Lewis Prince
IAzure Foundry MVP
AI Engineer
Welcome to The Data Rhino, my blog to discuss all things data that I involve myself in. This will be primarily be talking about AI through the Microsoft Stack.