Model Deployment & Serving Foundations

Expose trained models as HTTP APIs, handle payload parsing, and understand real-time vs. batch inference patterns.

Assumes: Experiment Tracking & MLflow.

Modules

1. Inference Paradigms (Real-time, Batch, Streaming)
2. Serialization Formats (Pickle, ONNX, TorchScript)
3. Anatomy of a Model Serving API

Tools

• Python
• REST APIs
• Linux
• Bash