Microsoft's skills measured for AI-300, mapped to the chapters of Ultra Transcenders AI-300: Operationalizing Machine Learning and Generative AI Solutions.
| Objective | Chapters |
|---|---|
| Design and implement an MLOps infrastructure (15–20%) | |
| Create and manage resources in a Machine Learning workspace | 1. Azure Machine Learning workspaces, assets and compute; 2. Automating and securing ML infrastructure |
| Create and manage assets in a Machine Learning workspace | 1. Azure Machine Learning workspaces, assets and compute; 3. Training models and tracking experiments |
| Implement IaC for Machine Learning | 2. Automating and securing ML infrastructure |
| Implement machine learning model lifecycle and operations (25–30%) | |
| Orchestrate model training | 3. Training models and tracking experiments |
| Implement model registration and versioning | 4. Registering models, responsible AI and lifecycle |
| Deploy machine learning models for production environments | 5. Deploying and monitoring machine learning models |
| Monitor and maintain machine learning models in production | 5. Deploying and monitoring machine learning models |
| Design and implement a GenAIOps infrastructure (20–25%) | |
| Implement Foundry environments and platform configuration | 6. The Microsoft Foundry platform: resources, identity and networking |
| Deploy and manage foundation models for production workloads | 7. Deploying foundation models |
| Implement prompt versioning and management with source control | 8. Prompt design, variants and version control |
| Implement generative AI quality assurance and observability (10–15%) | |
| Configure evaluation and validation for generative AI applications and agents | 9. Evaluation and observability for generative AI |
| Implement observability for generative AI applications and agents | 7. Deploying foundation models; 9. Evaluation and observability for generative AI |
| Optimize generative AI systems and model performance (10–15%) | |
| Optimize retrieval-augmented generation (RAG) performance and accuracy | 10. Optimising retrieval-augmented generation |
| Implement advanced fine-tuning and model customization | 11. Fine-tuning foundation models |
The full list of tasks under each objective is in Microsoft's official study guide. The book is organised by technology, so one chapter often serves several objectives.