AI Model Deployment

90 hours 6 Modules 10+ Hands-on Labs AI/ML Deployment

Course Description

This course develops practical and professional skills related to the deployment, testing, documentation, and monitoring of AI/ML models in production environments.

You will learn to build and ship complete AI-powered services using modern tools and best practices:

Topic	Description
Deployment Foundations	Scope definition, data dependencies, infrastructure planning
Model Training & Evaluation	Retraining pipelines, metrics, serialization (pickle, ONNX, joblib)
API Development	REST APIs with FastAPI/Flask, Swagger/OpenAPI documentation
AI-Assisted Coding	Prompt engineering, AI code generation, security considerations
Testing & Explainability	pytest, Postman, LIME, SHAP, model interpretability
End-to-End Project	Complete deployment lifecycle from training to production

Learning Objectives

By the end of this course, you will be able to:

Define the scope and requirements for deploying an AI model
Train and evaluate models using structured pipelines and rigorous metrics
Build and document production-ready REST APIs serving AI predictions
Leverage AI-assisted coding tools effectively and securely
Test APIs and models systematically with automated tools
Explain model predictions using interpretability frameworks (LIME, SHAP)
Deploy an end-to-end AI service with documentation and monitoring

Course Structure

Prerequisites

Course	Code	Description
Python Programming	420-XXX-BB	Variables, functions, OOP, data structures
Machine Learning Fundamentals	420-XXX-BB	Supervised/unsupervised learning, scikit-learn
Version Control (Git)	—	Branching, commits, pull requests

Hands-on Labs

Lab	Module	Objectives	Duration
TP1	Deployment Foundations	Define project scope, set up Python environment	60 min
TP2	Model Training	Train, evaluate, and serialize a model	90 min
TP3	FastAPI Basics	Build a prediction API with FastAPI	90 min
TP4	Flask API	Build an equivalent API with Flask	60 min
TP5	API Documentation	Generate Swagger/OpenAPI docs	45 min
TP6	AI-Assisted Coding	Generate and debug code with AI tools	60 min
TP7	Testing with pytest	Write unit and integration tests	60 min
TP8	API Testing with Postman	Build a Postman collection for your API	45 min
TP9	Model Explainability	Apply LIME and SHAP to your model	75 min
TP10	Final Project	Deploy a complete AI service end-to-end	180 min

Assessments

Week	Assessment	Weight	Content
3	Assessment 1	15%	Project brief + environment setup
5	Assessment 2	20%	Model evaluation report + serialization
8	Assessment 3	25%	Functional API service + documentation
12	Assessment 4	10%	AI coding reflection + debugging report
15	LIA Final Project	30%	Complete deployment + report + oral presentation

Technology Stack

Category	Tools
Language	Python 3.10+
ML Frameworks	scikit-learn, pandas, NumPy
API Frameworks	FastAPI, Flask
Testing	pytest, Postman, httpx
Explainability	LIME, SHAP
Serialization	pickle, joblib, ONNX
Documentation	Swagger/OpenAPI, Markdown
AI Tools	GitHub Copilot, ChatGPT, Cursor
DevOps	Docker, Git, virtual environments