Implement AI Solutions Using Azure Foundry
This domain covers 55-60% of the AI-901 exam — the heaviest section. You'll learn to use Microsoft Foundry to deploy models, build agents, work with multimodal capabilities, and implement AI solutions using the portal and SDK.
What you'll learn
- Microsoft Foundry basics: Portal navigation, catalog, Foundry SDK
- Model deployment: Deploy, configure, and use prebuilt and custom models
- Agents in Foundry: Build, configure, and test multi-turn chat apps and agents
- Content Understanding: Information extraction from documents and unstructured data
- Multimodal apps: Combining vision, audio, and text capabilities
- Prompt engineering: System vs user prompts, visual prompts, effective prompt construction
Azure Services Covered
| Service | Key Capabilities |
|---|---|
| Azure AI Foundry | Model catalog, hub/project, agents, evaluation, deployments |
| Azure OpenAI Service | GPT-4o, GPT-4, DALL-E, Whisper, embeddings |
| Azure AI Content Understanding | Document extraction, multimodal analysis |
Challenges
| # | Challenge | Focus |
|---|---|---|
| 10 | Image Classification | Computer vision workloads in Foundry |
| 11 | Object Detection | Detection models and use cases |
| 12 | Optical Character Recognition (OCR) | Text extraction from images |
| 13 | Face Detection and Analysis | Face API capabilities |
| 14 | Text Analytics: Key Phrases and Entities | NLP services |
| 15 | Sentiment Analysis and Language Detection | Language understanding |
| 16 | Speech Recognition and Synthesis | Speech services |
| 17 | Language Translation | Translator service |
| 18 | Azure AI Language and Speech Services | Service integration |
| 19 | Generative AI Fundamentals | LLMs, foundation models, GenAI concepts |
| 20 | Azure OpenAI Service | Models, deployments, playground |
| 21 | Azure AI Foundry | Platform, model catalog, hub/project |
| 22 | Prompt Engineering Basics | Techniques, temperature, grounding |
| 23 | Responsible AI for Generative AI | Content filters, safety, transparency |