@xdev-asia/xdev-knowledge-mcp 1.0.42 → 1.0.44

package/content/series/luyen-thi/luyen-thi-gcp-ml-engineer/chapters/04-phan-4-deployment-mlops/lessons/07-bai-7-model-deployment.md

@@ -0,0 +1,134 @@
+---
+id: 019c9619-lt03-l07
+title: 'Bài 7: Model Deployment & Prediction'
+slug: bai-7-model-deployment
+description: >-
+  Vertex AI Endpoints: online, batch prediction.
+  Model versioning, traffic splitting. Edge deployment.
+  Scaling config, GPU allocation.
+duration_minutes: 60
+is_free: true
+video_url: null
+sort_order: 7
+section_title: "Phần 4: Model Deployment & MLOps"
+course:
+  id: 019c9619-lt03-7003-c003-lt0300000003
+  title: 'Luyện thi Google Cloud Professional Machine Learning Engineer'
+  slug: luyen-thi-gcp-ml-engineer
+---
+
+<div style="text-align: center; margin: 2rem 0;">
+<img src="/storage/uploads/2026/04/gcp-mle-bai7-deployment.png" alt="Vertex AI Model Deployment" style="max-width: 800px; width: 100%; border-radius: 12px;" />
+<p><em>Vertex AI Deployment: Online Prediction, Batch Prediction, traffic splitting, và edge deployment</em></p>
+</div>
+
+<h2 id="deployment-types"><strong>1. Prediction Types on Vertex AI</strong></h2>
+
+<table>
+<thead><tr><th>Type</th><th>Latency</th><th>When to Use</th></tr></thead>
+<tbody>
+<tr><td><strong>Online Prediction</strong></td><td>Milliseconds (sync)</td><td>Real-time apps, user-facing APIs</td></tr>
+<tr><td><strong>Batch Prediction</strong></td><td>Minutes/Hours (async)</td><td>Large datasets, scheduled scoring</td></tr>
+<tr><td><strong>Streaming Prediction</strong></td><td>Near real-time</td><td>Pub/Sub events + Dataflow + Vertex AI</td></tr>
+</tbody>
+</table>
+
+<h2 id="vertex-endpoints"><strong>2. Vertex AI Endpoints</strong></h2>
+
+<pre><code class="language-text">Vertex AI Endpoint Architecture:
+
+Client Request
+    ↓
+Vertex AI Endpoint (load balancer)
+    ├── Model Version A (70% traffic)
+    │       └── Deployed Model (e.g., v1.0)
+    └── Model Version B (30% traffic)  ← Canary/A-B test
+            └── Deployed Model (e.g., v1.1)
+</code></pre>
+
+<p>Mỗi Endpoint có thể có <strong>nhiều model versions</strong> với <strong>traffic splitting</strong> — dùng để A/B testing và canary deployments.</p>
+
+<table>
+<thead><tr><th>Feature</th><th>Details</th></tr></thead>
+<tbody>
+<tr><td><strong>Dedicated Endpoint</strong></td><td>Dedicated resources, lowest latency, higher cost</td></tr>
+<tr><td><strong>Shared Endpoint</strong></td><td>Multi-tenant, lower cost, potential cold start</td></tr>
+<tr><td><strong>Explanation</strong></td><td>Enable Vertex Explainability per deployed model</td></tr>
+<tr><td><strong>Min/Max Replicas</strong></td><td>Autoscaling based on request rate</td></tr>
+<tr><td><strong>GPU allocation</strong></td><td>Specify GPU type (NVIDIA T4, A100) per deployment</td></tr>
+</tbody>
+</table>
+
+<blockquote>
+<p><strong>Exam tip:</strong> Traffic splitting trong Vertex AI Endpoints là cách triển khai <strong>Canary deployment</strong> hoặc <strong>A/B testing</strong>. Câu hỏi "roll out new model version safely" → Traffic splitting (ví dụ: 90% old, 10% new).</p>
+</blockquote>
+
+<h2 id="batch-prediction"><strong>3. Batch Prediction</strong></h2>
+
+<table>
+<thead><tr><th>Property</th><th>Value</th></tr></thead>
+<tbody>
+<tr><td><strong>Input</strong></td><td>Cloud Storage (CSV, JSON, JSONL, TFRecords, Avro)</td></tr>
+<tr><td><strong>Output</strong></td><td>Cloud Storage (predictions as JSON/CSV)</td></tr>
+<tr><td><strong>No Endpoint needed</strong></td><td>Runs directly from Model Registry, no persistent endpoint</td></tr>
+<tr><td><strong>Auto-scaling</strong></td><td>Scales to zero when done (cost-efficient)</td></tr>
+<tr><td><strong>Accelerators</strong></td><td>Supports GPU/TPU for batch inference</td></tr>
+</tbody>
+</table>
+
+<h2 id="model-versioning"><strong>4. Model Versioning & Registry</strong></h2>
+
+<pre><code class="language-text">Vertex AI Model Registry:
+
+Model: churn-predictor
+├── v1 (Logistic Regression)  ← Champion in production
+│   - Accuracy: 0.87
+│   - Deployed to: endpoint/prod (70% traffic)
+│
+└── v2 (XGBoost)              ← Challenger
+    - Accuracy: 0.91
+    - Deployed to: endpoint/prod (30% traffic)
+
+After validation: promote v2 to Champion
+</code></pre>
+
+<h2 id="edge-deployment"><strong>5. Edge Deployment</strong></h2>
+
+<table>
+<thead><tr><th>Platform</th><th>Solution</th></tr></thead>
+<tbody>
+<tr><td>Mobile (Android/iOS)</td><td>TFLite + Vertex AI model export</td></tr>
+<tr><td>Edge devices (IoT)</td><td>TFLite Micro / Edge TPU (Coral)</td></tr>
+<tr><td>On-premise servers</td><td>TF Serving in Docker container</td></tr>
+<tr><td>Kubernetes</td><td>KServe (formerly KFServing) on GKE</td></tr>
+</tbody>
+</table>
+
+<h2 id="practice"><strong>6. Practice Questions</strong></h2>
+
+<p><strong>Q1:</strong> A company needs to score 50 million customer records for churn risk. Results are needed within 2 hours but not in real time. Which Vertex AI prediction option is MOST cost-effective?</p>
+<ul>
+<li>A) Online Prediction with high replica count</li>
+<li>B) Batch Prediction ✓</li>
+<li>C) Streaming prediction via Dataflow</li>
+<li>D) Deploy on dedicated GPU endpoint</li>
+</ul>
+<p><em>Explanation: Batch Prediction is designed for large-scale asynchronous scoring. It scales compute resources up during the job and back to zero when done, with no persistent endpoint cost. Online Prediction would be wasteful since real-time response isn't needed for batch scoring.</em></p>
+
+<p><strong>Q2:</strong> A team is deploying a new model version. They want to gradually route 10% of production traffic to the new version while the old version handles 90%, allowing comparison of performance metrics before full rollout. Which Vertex AI feature enables this?</p>
+<ul>
+<li>A) Model Registry versioning</li>
+<li>B) Traffic splitting on Vertex AI Endpoints ✓</li>
+<li>C) Batch Prediction comparison</li>
+<li>D) Vertex AI Experiments</li>
+</ul>
+<p><em>Explanation: Vertex AI Endpoints support deploying multiple model versions simultaneously with configurable traffic splits (e.g., 90%/10%). This enables canary deployments and A/B testing to compare live performance before committing to a full rollout.</em></p>
+
+<p><strong>Q3:</strong> A retail company wants to detect product defects on a factory floor without network connectivity to cloud. Which deployment approach should they use?</p>
+<ul>
+<li>A) Vertex AI Online Prediction Endpoint</li>
+<li>B) AutoML Edge Model deployed to device using TFLite ✓</li>
+<li>C) BigQuery ML batch prediction</li>
+<li>D) TF Serving on Cloud Run</li>
+</ul>
+<p><em>Explanation: Edge deployment with TFLite (or AutoML Edge Model) runs inference locally on the device without network connectivity. TFLite supports on-device inference for computer vision models, suitable for factory floor equipment with no internet access.</em></p>

package/content/series/luyen-thi/luyen-thi-gcp-ml-engineer/chapters/04-phan-4-deployment-mlops/lessons/08-bai-8-vertex-ai-pipelines-mlops.md

@@ -0,0 +1,149 @@
+---
+id: 019c9619-lt03-l08
+title: 'Bài 8: Vertex AI Pipelines & MLOps'
+slug: bai-8-vertex-ai-pipelines-mlops
+description: >-
+  Vertex AI Pipelines (Kubeflow Pipelines SDK).
+  Model Registry, Experiments, Metadata Store.
+  Vertex AI Model Monitoring: skew, drift detection.
+  CI/CD cho ML: Cloud Build + Vertex AI.
+duration_minutes: 60
+is_free: true
+video_url: null
+sort_order: 8
+section_title: "Phần 4: Model Deployment & MLOps"
+course:
+  id: 019c9619-lt03-7003-c003-lt0300000003
+  title: 'Luyện thi Google Cloud Professional Machine Learning Engineer'
+  slug: luyen-thi-gcp-ml-engineer
+---
+
+<div style="text-align: center; margin: 2rem 0;">
+<img src="/storage/uploads/2026/04/gcp-mle-bai8-mlops-cicd.png" alt="Vertex AI Pipelines & MLOps" style="max-width: 800px; width: 100%; border-radius: 12px;" />
+<p><em>Vertex AI MLOps: Pipelines, CI/CD, Model Registry, và monitoring cho production ML</em></p>
+</div>
+
+<h2 id="mlops-maturity"><strong>1. MLOps Maturity Levels</strong></h2>
+
+<table>
+<thead><tr><th>Level</th><th>Description</th><th>Automation</th></tr></thead>
+<tbody>
+<tr><td><strong>Level 0</strong></td><td>Manual process, scripts only</td><td>None</td></tr>
+<tr><td><strong>Level 1</strong></td><td>ML pipeline automation, continuous training</td><td>Training pipeline</td></tr>
+<tr><td><strong>Level 2</strong></td><td>Full CI/CD for ML, automated retraining triggers</td><td>Everything</td></tr>
+</tbody>
+</table>
+
+<h2 id="vertex-pipelines"><strong>2. Vertex AI Pipelines</strong></h2>
+
+<p>Vertex AI Pipelines là managed execution environment cho <strong>Kubeflow Pipelines (KFP)</strong>. Pipeline được định nghĩa bằng Python SDK và compile thành YAML.</p>
+
+<pre><code class="language-text">Vertex AI Pipeline Structure:
+
+@component (preprocess_data)
+     ↓
+@component (train_model)
+     ↓
+@component (evaluate_model)
+     ↓ (if accuracy > threshold)
+@component (deploy_model)
+
+Each component = isolated Docker container
+Artifacts (data, models) stored in Cloud Storage
+Metadata tracked in Vertex ML Metadata Store
+</code></pre>
+
+<table>
+<thead><tr><th>Pipeline SDK</th><th>Notes</th></tr></thead>
+<tbody>
+<tr><td><strong>Kubeflow Pipelines SDK v2</strong></td><td>Primary SDK for Vertex AI Pipelines</td></tr>
+<tr><td><strong>TFX</strong></td><td>TensorFlow-specific pipeline components</td></tr>
+<tr><td><strong>Google Cloud Pipeline Components</strong></td><td>Pre-built components cho Vertex AI services</td></tr>
+</tbody>
+</table>
+
+<h2 id="model-monitoring"><strong>3. Vertex AI Model Monitoring</strong></h2>
+
+<table>
+<thead><tr><th>Monitoring Type</th><th>What It Detects</th></tr></thead>
+<tbody>
+<tr><td><strong>Feature Skew Monitoring</strong></td><td>Serving feature distribution ≠ training baseline</td></tr>
+<tr><td><strong>Feature Drift Monitoring</strong></td><td>Serving feature distribution changes over time</td></tr>
+<tr><td><strong>Prediction Drift</strong></td><td>Model output distribution changes (indirect label drift)</td></tr>
+</tbody>
+</table>
+
+<pre><code class="language-text">Model Monitoring Workflow:
+
+Training Data Baseline (BigQuery/GCS)
+     ↓ (establish distribution)
+Deploy to Endpoint with Monitoring enabled
+     ↓ (collect serving requests)
+Periodic Analysis (hourly/daily)
+     ↓ (compare distributions)
+Alert if skew/drift > threshold
+     ↓
+Retrain trigger → new Pipeline run
+</code></pre>
+
+<h2 id="experiments-metadata"><strong>4. Vertex AI Experiments & Metadata</strong></h2>
+
+<table>
+<thead><tr><th>Component</th><th>Purpose</th></tr></thead>
+<tbody>
+<tr><td><strong>Vertex AI Experiments</strong></td><td>Track hyperparameters, metrics, artifacts across runs</td></tr>
+<tr><td><strong>ML Metadata Store</strong></td><td>Track lineage: data → model → endpoint</td></tr>
+<tr><td><strong>Vertex AI TensorBoard</strong></td><td>Visualize training metrics (loss, accuracy curves)</td></tr>
+</tbody>
+</table>
+
+<h2 id="cicd-ml"><strong>5. CI/CD for ML on GCP</strong></h2>
+
+<pre><code class="language-text">ML CI/CD Pipeline on GCP:
+
+Code Push to Cloud Source Repositories
+     ↓
+Cloud Build trigger (CI)
+     ├── Unit tests for ML components
+     ├── Data validation tests
+     └── Build Docker image → push to Artifact Registry
+          ↓
+Vertex AI Pipeline trigger (CD/CT)
+     ├── Data preprocessing
+     ├── Model training
+     ├── Model evaluation
+     └── Conditional deployment → Vertex AI Endpoint
+</code></pre>
+
+<blockquote>
+<p><strong>Exam tip:</strong> CI/CD cho ML = Cloud Build (code testing + Docker build) + Vertex AI Pipelines (training + deployment orchestration). Cloud Source Repositories là GCP's Git hosting. Artifact Registry thay thế Container Registry để lưu Docker images.</p>
+</blockquote>
+
+<h2 id="practice"><strong>6. Practice Questions</strong></h2>
+
+<p><strong>Q1:</strong> A production ML model's prediction distribution has shifted significantly over 3 weeks, but ground truth labels are not yet available to measure accuracy directly. Which Vertex AI monitoring type detects this?</p>
+<ul>
+<li>A) Feature Skew Monitoring</li>
+<li>B) Prediction Drift Monitoring ✓</li>
+<li>C) Training data validation</li>
+<li>D) Vertex AI Experiments baseline comparison</li>
+</ul>
+<p><em>Explanation: Prediction Drift Monitoring tracks how the model's output distribution changes over time, serving as an indirect signal of model degradation even when ground truth labels are unavailable. Feature Skew compares serving vs training feature distributions (requires known training baseline).</em></p>
+
+<p><strong>Q2:</strong> A team is building a Vertex AI Pipeline that includes data preprocessing, model training, and deployment. They need to track all inputs, outputs, and model artifacts for auditability and reproducibility. Which service stores this lineage information?</p>
+<ul>
+<li>A) Cloud Logging</li>
+<li>B) Vertex AI ML Metadata Store ✓</li>
+<li>C) Cloud Storage versioning</li>
+<li>D) Vertex AI Experiments dashboard</li>
+</ul>
+<p><em>Explanation: Vertex AI ML Metadata Store (also called Vertex ML Metadata) automatically tracks lineage: which datasets produced which models, which models were deployed to which endpoints, including hyperparameters and evaluation metrics — enabling full provenance tracking.</em></p>
+
+<p><strong>Q3:</strong> A company wants to automatically retrain their ML model whenever new training data is available in Cloud Storage. The retraining should run a Vertex AI Pipeline and deploy if metrics pass thresholds. Which GCP service should trigger the pipeline?</p>
+<ul>
+<li>A) Vertex AI Schedules</li>
+<li>B) Cloud Storage notifications + Cloud Functions/Eventarc → Vertex AI Pipelines ✓</li>
+<li>C) BigQuery scheduled queries</li>
+<li>D) Cloud Scheduler alone</li>
+</ul>
+<p><em>Explanation: Cloud Storage object finalize notifications can trigger Cloud Functions or Eventarc, which then programmatically start a Vertex AI Pipeline run. This creates event-driven continuous training (MLOps Level 1). Cloud Scheduler triggers on time, not on data availability.</em></p>

package/content/series/luyen-thi/luyen-thi-gcp-ml-engineer/chapters/05-phan-5-responsible-ai/lessons/09-bai-9-responsible-ai.md

@@ -0,0 +1,128 @@
+---
+id: 019c9619-lt03-l09
+title: 'Bài 9: Responsible AI & Security'
+slug: bai-9-responsible-ai
+description: >-
+  Google Responsible AI principles. Vertex AI Explainability (SHAP, IG).
+  Fairness indicators. Privacy: differential privacy, federated learning.
+  IAM, VPC-SC, CMEK cho ML workloads.
+duration_minutes: 50
+is_free: true
+video_url: null
+sort_order: 9
+section_title: "Phần 5: Responsible AI & Ôn tập"
+course:
+  id: 019c9619-lt03-7003-c003-lt0300000003
+  title: 'Luyện thi Google Cloud Professional Machine Learning Engineer'
+  slug: luyen-thi-gcp-ml-engineer
+---
+
+<h2 id="responsible-ai"><strong>1. Google's Responsible AI Principles</strong></h2>
+
+<table>
+<thead><tr><th>Principle</th><th>Key Requirement</th></tr></thead>
+<tbody>
+<tr><td><strong>Socially Beneficial</strong></td><td>Benefits society and individuals</td></tr>
+<tr><td><strong>Avoid Unfair Bias</strong></td><td>Test fairness across demographic groups</td></tr>
+<tr><td><strong>Safety</strong></td><td>Test across diverse scenarios, continuous evaluation</td></tr>
+<tr><td><strong>Accountable</strong></td><td>Appropriate human oversight and control</td></tr>
+<tr><td><strong>Privacy Preserving</strong></td><td>Protect training data privacy</td></tr>
+<tr><td><strong>Scientific Excellence</strong></td><td>Rigorous research standards</td></tr>
+<tr><td><strong>Available for Beneficial Uses</strong></td><td>Primary benefit criteria</td></tr>
+</tbody>
+</table>
+
+<h2 id="explainability"><strong>2. Vertex AI Explainability</strong></h2>
+
+<p>Vertex AI Explainability cung cấp feature attribution scores — giải thích tại sao model đưa ra prediction nào đó.</p>
+
+<table>
+<thead><tr><th>Method</th><th>For</th><th>How</th></tr></thead>
+<tbody>
+<tr><td><strong>SHAP (Shapley Values)</strong></td><td>Tabular models</td><td>Game theory: contribution của mỗi feature</td></tr>
+<tr><td><strong>Integrated Gradients (IG)</strong></td><td>Neural networks (image, text)</td><td>Gradient accumulation from baseline to input</td></tr>
+<tr><td><strong>XRAI</strong></td><td>Image models</td><td>Pixel-region attribution (better UX than IG)</td></tr>
+<tr><td><strong>Sampled Shapley</strong></td><td>Large tabular datasets</td><td>Approximate SHAP, faster</td></tr>
+</tbody>
+</table>
+
+<blockquote>
+<p><strong>Exam tip:</strong> "Explain why a loan was denied" → SHAP for tabular models. "Highlight which image regions drove classification" → Integrated Gradients or XRAI. Vertex AI Explainability phải được enable lúc deploy endpoint.</p>
+</blockquote>
+
+<h2 id="fairness"><strong>3. Fairness & Bias Detection</strong></h2>
+
+<table>
+<thead><tr><th>Tool/Concept</th><th>Description</th></tr></thead>
+<tbody>
+<tr><td><strong>Fairness Indicators</strong></td><td>GCP tool: evaluate model fairness metrics across demographic slices</td></tr>
+<tr><td><strong>What-If Tool</strong></td><td>Interactive exploration of model behavior, counterfactuals</td></tr>
+<tr><td><strong>Demographic parity</strong></td><td>Model predicts same rate across demographic groups</td></tr>
+<tr><td><strong>Equal opportunity</strong></td><td>Same recall/TPR across groups</td></tr>
+<tr><td><strong>Data slice evaluation</strong></td><td>Evaluate metrics per gender, race, age in TFX Evaluator</td></tr>
+</tbody>
+</table>
+
+<h2 id="privacy"><strong>4. Privacy Techniques</strong></h2>
+
+<table>
+<thead><tr><th>Technique</th><th>Description</th></tr></thead>
+<tbody>
+<tr><td><strong>Differential Privacy</strong></td><td>Add statistical noise to training data/model, prevents individual data re-identification</td></tr>
+<tr><td><strong>Federated Learning</strong></td><td>Train on distributed data without centralizing raw data — model updates only</td></tr>
+<tr><td><strong>Data Anonymization</strong></td><td>Remove PII before training (Cloud DLP API)</td></tr>
+</tbody>
+</table>
+
+<h2 id="security"><strong>5. Security Controls for ML Workloads</strong></h2>
+
+<table>
+<thead><tr><th>Control</th><th>Purpose</th></tr></thead>
+<tbody>
+<tr><td><strong>IAM roles</strong></td><td>Least-privilege access for ML service accounts</td></tr>
+<tr><td><strong>VPC Service Controls (VPC-SC)</strong></td><td>Security perimeter: prevent data exfiltration from BigQuery, GCS</td></tr>
+<tr><td><strong>CMEK (Customer-Managed Encryption Keys)</strong></td><td>Control encryption keys via Cloud KMS</td></tr>
+<tr><td><strong>Private IP for Vertex AI</strong></td><td>Training and endpoints use private networking</td></tr>
+<tr><td><strong>Cloud Audit Logs</strong></td><td>Who accessed what data, when (Data Access + Admin Activity)</td></tr>
+</tbody>
+</table>
+
+<pre><code class="language-text">VPC Service Controls Perimeter:
+
+┌────── Security Perimeter ─────────┐
+│  BigQuery  │  Cloud Storage       │
+│  Vertex AI │  Cloud KMS           │
+│  Dataflow  │  Secret Manager      │
+└──────────────────────────────────┘
+         │ (no exfiltration outside perimeter)
+         ✗ Unauthorized access blocked
+</code></pre>
+
+<h2 id="practice"><strong>6. Practice Questions</strong></h2>
+
+<p><strong>Q1:</strong> A financial services company deployed a loan approval ML model. Regulators require the company to explain why specific loan applications were denied. Which Vertex AI feature provides per-prediction feature importance scores for tabular models?</p>
+<ul>
+<li>A) Vertex AI Experiments</li>
+<li>B) Vertex AI Explainability with SHAP ✓</li>
+<li>C) Vertex AI Model Monitoring</li>
+<li>D) Fairness Indicators</li>
+</ul>
+<p><em>Explanation: Vertex AI Explainability with Shapley Values (SHAP) assigns an importance score to each feature for each individual prediction, explaining why a specific loan was denied by attributing the model's decision to specific input features like credit_score, income, debt_ratio.</em></p>
+
+<p><strong>Q2:</strong> A healthcare company needs to train ML models on patient data distributed across multiple hospitals. Data privacy regulations prohibit centralizing raw patient records. Which privacy-preserving ML approach should they use?</p>
+<ul>
+<li>A) Differential Privacy with central training</li>
+<li>B) Federated Learning ✓</li>
+<li>C) Data anonymization + BigQuery ML</li>
+<li>D) Cloud DLP de-identification</li>
+</ul>
+<p><em>Explanation: Federated Learning trains models on distributed data without moving raw data to a central location. Each hospital trains locally on its own data; only model updates (gradients) are shared and aggregated. Raw patient records never leave the hospital's environment.</em></p>
+
+<p><strong>Q3:</strong> A company processes sensitive financial data in BigQuery for ML training. They need to prevent data from being moved outside an approved security boundary to unauthorized GCP projects. Which GCP feature should they implement?</p>
+<ul>
+<li>A) Cloud KMS CMEK encryption</li>
+<li>B) VPC Service Controls (VPC-SC) perimeter ✓</li>
+<li>C) IAM role deny policies</li>
+<li>D) Cloud Armor WAF</li>
+</ul>
+<p><em>Explanation: VPC Service Controls creates a security perimeter around GCP services (BigQuery, Cloud Storage, Vertex AI). It prevents data exfiltration by blocking requests that would move data outside the defined perimeter, even from authenticated users. CMEK provides encryption control but doesn't prevent exfiltration.</em></p>

package/content/series/luyen-thi/luyen-thi-gcp-ml-engineer/chapters/05-phan-5-responsible-ai/lessons/10-bai-10-cheat-sheet-chien-luoc-thi.md

@@ -0,0 +1,108 @@
+---
+id: 019c9619-lt03-l10
+title: 'Bài 10: Cheat Sheet & Chiến lược thi GCP MLE'
+slug: bai-10-cheat-sheet-chien-luoc-thi
+description: >-
+  Bảng tổng hợp toàn khoá GCP Professional Machine Learning Engineer.
+  GCP service reference, evaluation metrics, domain weights, và chiến lược thi.
+duration_minutes: 40
+is_free: true
+video_url: null
+sort_order: 10
+section_title: "Phần 5: Responsible AI & Ôn tập"
+course:
+  id: 019c9619-lt03-7003-c003-lt0300000003
+  title: 'Luyện thi Google Cloud Professional Machine Learning Engineer'
+  slug: luyen-thi-gcp-ml-engineer
+---
+
+<h2 id="exam-structure"><strong>1. Cấu Trúc Đề Thi GCP Professional ML Engineer</strong></h2>
+
+<table>
+<thead><tr><th>Item</th><th>Details</th></tr></thead>
+<tbody>
+<tr><td><strong>Total Questions</strong></td><td>60 câu</td></tr>
+<tr><td><strong>Time Limit</strong></td><td>120 phút (2 giờ)</td></tr>
+<tr><td><strong>Passing Score</strong></td><td>~70% (Google không công bố chính xác)</td></tr>
+<tr><td><strong>Format</strong></td><td>Multiple choice, multiple select</td></tr>
+<tr><td><strong>Validity</strong></td><td>2 năm</td></tr>
+<tr><td><strong>Level</strong></td><td>Professional (intermediate to advanced)</td></tr>
+</tbody>
+</table>
+
+<h2 id="domain-weights"><strong>2. Domain Weights</strong></h2>
+
+<table>
+<thead><tr><th>Domain</th><th>Weight</th></tr></thead>
+<tbody>
+<tr><td>1. Architecting low-code ML solutions</td><td>~10%</td></tr>
+<tr><td>2. Collaborate within and across teams to manage data and models</td><td>~20%</td></tr>
+<tr><td>3. Scale prototypes into ML models</td><td>~20%</td></tr>
+<tr><td>4. Serve and scale models</td><td>~20%</td></tr>
+<tr><td>5. Automate & orchestrate ML pipelines</td><td>~20%</td></tr>
+<tr><td>6. Monitor ML solutions</td><td>~10%</td></tr>
+</tbody>
+</table>
+
+<h2 id="service-cheat-sheet"><strong>3. GCP ML Services Cheat Sheet</strong></h2>
+
+<table>
+<thead><tr><th>Task</th><th>GCP Service</th></tr></thead>
+<tbody>
+<tr><td>No-code image classification</td><td>Vertex AI AutoML Image</td></tr>
+<tr><td>SQL-based ML in data warehouse</td><td>BigQuery ML</td></tr>
+<tr><td>Custom TensorFlow/PyTorch training</td><td>Vertex AI Custom Training</td></tr>
+<tr><td>Hyperparameter optimization</td><td>Vertex AI Hyperparameter Tuning (Bayesian)</td></tr>
+<tr><td>Feature consistency training/serving</td><td>Vertex AI Feature Store</td></tr>
+<tr><td>ML workflow orchestration (pipelines)</td><td>Vertex AI Pipelines (KFP)</td></tr>
+<tr><td>Experiment tracking</td><td>Vertex AI Experiments</td></tr>
+<tr><td>Model versioning</td><td>Vertex AI Model Registry</td></tr>
+<tr><td>A/B testing model versions</td><td>Vertex AI Endpoints traffic splitting</td></tr>
+<tr><td>Monitor feature skew/drift</td><td>Vertex AI Model Monitoring</td></tr>
+<tr><td>Explain model predictions</td><td>Vertex AI Explainability (SHAP, IG)</td></tr>
+<tr><td>Real-time event ingestion</td><td>Pub/Sub</td></tr>
+<tr><td>Batch + streaming ETL (unified)</td><td>Dataflow (Apache Beam)</td></tr>
+<tr><td>Spark/Hadoop workloads</td><td>Dataproc</td></tr>
+<tr><td>ML pipeline orchestration (multi-service)</td><td>Cloud Composer (Airflow)</td></tr>
+<tr><td>Natural language analysis (no training)</td><td>Cloud Natural Language API</td></tr>
+<tr><td>Document extraction</td><td>Document AI</td></tr>
+<tr><td>Speech to text</td><td>Cloud Speech-to-Text API</td></tr>
+<tr><td>Prevent data exfiltration</td><td>VPC Service Controls</td></tr>
+<tr><td>Customer-managed encryption</td><td>Cloud KMS (CMEK)</td></tr>
+</tbody>
+</table>
+
+<h2 id="traps"><strong>4. Common Exam Traps</strong></h2>
+
+<table>
+<thead><tr><th>Trap</th><th>Correct Answer</th></tr></thead>
+<tbody>
+<tr><td>"No ML expertise, image classification"</td><td>AutoML Image (not custom training)</td></tr>
+<tr><td>"Train on data already in BigQuery"</td><td>BigQuery ML (not Vertex AI)</td></tr>
+<tr><td>"Features differ at training vs serving"</td><td>Vertex AI Feature Store (not re-training)</td></tr>
+<tr><td>"Trigger retraining when data arrives"</td><td>GCS notification → Eventarc → Vertex AI Pipeline</td></tr>
+<tr><td>"Explain why model rejected application"</td><td>Vertex AI Explainability (SHAP)</td></tr>
+<tr><td>"Train on distributed hospital data"</td><td>Federated Learning</td></tr>
+<tr><td>"Prevent BigQuery data exfiltration"</td><td>VPC Service Controls</td></tr>
+<tr><td>"Compare model performance across runs"</td><td>Vertex AI Experiments</td></tr>
+</tbody>
+</table>
+
+<blockquote>
+<p><strong>Exam tip:</strong> GCP Professional ML Engineer thường hỏi về architecture decisions, không phải API syntax. Key question patterns: "which service BEST fits the requirement", "what is the FIRST step", "which approach requires the LEAST operational overhead". Luôn ưu tiên managed services của GCP khi câu hỏi có "minimal management" hoặc "serverless".</p>
+</blockquote>
+
+<h2 id="study-plan"><strong>5. Kế Hoạch Ôn Tập</strong></h2>
+
+<table>
+<thead><tr><th>Ngày</th><th>Focus</th></tr></thead>
+<tbody>
+<tr><td>Day 1</td><td>Vertex AI full platform: Training, Pipelines, Endpoints, Monitoring</td></tr>
+<tr><td>Day 2</td><td>Data engineering: Pub/Sub, Dataflow, Dataproc, Cloud Composer</td></tr>
+<tr><td>Day 3</td><td>BigQuery ML + Feature Engineering + Feature Store</td></tr>
+<tr><td>Day 4</td><td>Responsible AI: Explainability, Fairness, Privacy, Security</td></tr>
+<tr><td>Day 5</td><td>Practice exam 1 — identify weak areas</td></tr>
+<tr><td>Day 6</td><td>Review weak areas + Practice exam 2</td></tr>
+<tr><td>Day 7</td><td>Cheat sheet review only</td></tr>
+</tbody>
+</table>

package/content/series/luyen-thi/luyen-thi-gcp-ml-engineer/index.md

@@ -6,7 +6,7 @@ description: >-
   Lộ trình ôn tập toàn diện cho kỳ thi Google Cloud Professional Machine Learning
   Engineer. Vertex AI, BigQuery ML, TFX pipeline, MLOps trên GCP.
 
-featured_image: null
+featured_image: images/blog/gcp-ml-engineer-series-banner.png
 level: advanced
 duration_hours: 35
 lesson_count: 10