# ML Engineering & Deployment

- **Part of:** [AI & Machine Learning](/en/tracks/ai-machine-learning) — learning path 3 of 3
- **Length:** 3 chapters · 14 lessons · 8 hours a week
- **Level:** Intermediate

::route
1. [Python & Mathematics for ML](/en/paths/python-mathematics-for-ml) — 4 chapters · 10 hours a week · 1 checkpoint
2. [Machine Learning Fundamentals](/en/paths/machine-learning-fundamentals) — 4 chapters · 12 hours a week · 1 checkpoint
3. **ML Engineering & Deployment** — 3 chapters · 8 hours a week · 1 checkpoint
::

::cta https://app.artimend.com/tracks/ai-machine-learning Start this track

## Syllabus

### 1. ML Project Structure & Experiment Tracking

- 5 lessons
- **Skills:** Experiment Tracking · MLflow · Weights & Biases · DVC · Project Structure

### 2. Model Serving & APIs

- 5 lessons
- **Skills:** FastAPI · Model Serving · Docker · Batch vs Online Inference

### 3. Monitoring, Drift & Retraining

- 4 lessons
- **Skills:** Model Monitoring · Data Drift Detection · Retraining Pipelines
- **Checkpoint:** MLOps Capstone Gate

## Checkpoints

To pass each checkpoint you hand in a piece of real work. What each one asks for:

- **MLOps Capstone Gate:** Ship a trained model behind a FastAPI / Flask endpoint with input validation and structured logging. Add a drift-detection script that flags when incoming feature distributions diverge from training.