Domain 6: Monitoring AI Solutions¶
Official exam weight: approximately 13%
Official objectives¶
6.1 Identify risks to AI solutions¶
- Protect systems from data or model leaks and unintentional exploitation, including data exfiltration, malicious prompting, and sharing sensitive data with LLMs.
- Understand controls such as regular expressions, safety filters, and Model Armor.
- Align with responsible AI practices, including monitoring for bias.
- Understand model explainability on Agent Platform, including Agent Platform Inference.
6.2 Monitor, test, and troubleshoot AI solutions¶
- Configure Model Monitoring on Gemini Enterprise Agent Platform for continuous evaluation metrics on production models.
- Monitor training-serving skew, data drift, concept drift, and feature attribution drift.
- Monitor, test, and evaluate generative AI solutions.
Guided labs to build¶
| Status | Lab | Objective |
|---|---|---|
| Planned | Identify training-serving skew and data drift | 6.2 |
| Planned | Distinguish data drift from concept drift | 6.2 |
| Planned | Configure model monitoring and alerting | 6.2 |
| Planned | Inspect feature attribution drift | 6.2 |
| Planned | Evaluate model explanations and bias | 6.1 |
| Planned | Protect a generative AI workflow from prompt and data risks | 6.1 |
| Planned | Evaluate and troubleshoot a generative AI application | 6.2 |
Google Skills¶
Open the official PMLE learning path for the current model monitoring, evaluation, and responsible AI activities.
Exam focus¶
Be able to distinguish training-serving skew, data drift, concept drift, and feature attribution drift. Also understand that AI monitoring includes security, responsible AI, explainability, and generative AI evaluation—not only infrastructure metrics.
📬 DevopsPilot Weekly — Learn DevOps, Cloud & Gen AI the simple way.
👉 Subscribe here