Skip to content

MLOps Course Repository – Complete Learning Guide

Welcome to the MLOps Course Repo, a hands-on, step-by-step repository designed for students and beginners to advanced learners to build a full production-grade MLOps pipeline.

This repository covers real-world tools, secure practices, and best practices used in the industry.

🚀 Course Roadmap

This course takes you from the basics of ML and DevOps to full production-grade MLOps:

  1. Environment Setup – Python, Docker, Kind, MLflow
  2. Data Management – Raw & processed datasets, version control with DVC
  3. Source Code & Pipelines – Modular ML pipelines: data ingestion, preprocessing, features, training, evaluation, prediction
  4. Experiment Tracking – MLflow experiments and reproducibility
  5. Model Serving – REST API with FastAPI, serverless with KServe and Seldon
  6. Testing & CI/CD – Unit tests, integration tests, GitHub Actions
  7. Orchestration – Airflow DAGs & Kubeflow pipelines
  8. Deployment – Kubernetes manifests, autoscaling, secure namespaces & RBAC
  9. GitOps – Continuous deployment using ArgoCD
  10. Monitoring – Infrastructure & application monitoring with Prometheus & Grafana
  11. Model Monitoring – Drift detection with Evidently AI
  12. Feature Store – Production-grade features using Feast
  13. Infrastructure as Code – Terraform for AWS EKS, S3, IAM, and KMS
  14. End-to-End Projects – Resume-ready, fully functional ML pipelines

📂 Folder Structure & Learning Index

Folder Description Learning Outcomes
01-setup Environment setup for Python, Docker, Kind, MLflow Learn to create a local dev environment and containerized Kubernetes cluster
02-data Raw & processed datasets Data management, preprocessing, and DVC versioning
03-src ML source code Data ingestion, validation, preprocessing, feature engineering, training, evaluation, prediction
03-src/01-data Data ingestion & validation Use Great Expectations for data quality checks
03-src/02-features Feature engineering Integrate with Feast Feature Store
03-src/03-models Training & evaluation Log metrics & artifacts using MLflow
03-src/04-pipelines Pipeline scripts Modular pipelines for local, Airflow, and Kubeflow execution
03-src/05-utils Utility scripts Logging, configuration, security (secrets management, encryption)
04-notebooks EDA & prototyping Jupyter notebooks for exploring datasets & features
05-dvc DVC pipelines Version datasets and maintain reproducibility
06-experiments MLflow experiment tracking Track parameters, metrics, artifacts, and compare models
07-serving Model serving Build production-ready REST APIs with FastAPI, KServe, and Seldon
08-tests Unit & integration tests Test data processing, models, pipelines, and APIs
09-ci-cd GitHub Actions workflows Automate training, testing, and deployment
10-k8s Kubernetes manifests Deploy models securely with Namespaces, RBAC, Network Policies, HPA autoscaling
11-kubeflow Kubeflow pipelines Orchestrate ML workflows for production
12-gitops ArgoCD GitOps deployment Automate deployment using GitOps practices
13-monitoring Prometheus & Grafana Monitor infrastructure and applications
14-monitoring-ml Model monitoring & drift detection Detect data/model drift and generate reports using Evidently AI
15-feature-store Feast Feature Store Manage and serve production-grade features
16-terraform AWS infrastructure as code Provision EKS cluster, S3, IAM, KMS securely
17-projects End-to-end ML projects Hands-on projects for practice and portfolio building

âš¡ Quick Start Guide

Bash
1
2
3
4
5
6
7
8
9
# Clone repository
git clone https://github.com/username/mlops.git
cd mlops

# Setup Python environment
bash 01-setup/01-python/setup-venv.sh

# Optional: Start MLflow tracking server
bash 01-setup/04-mlflow/start-mlflow.sh

Follow the folders in numerical order for a sequential, hands-on learning experience.

🛠 Tools & Security Features

Layer Tools / Practices
ML Tracking MLflow
Data Versioning DVC
Feature Store Feast
Orchestration Airflow, Kubeflow
Model Serving FastAPI, KServe, Seldon
CI/CD GitHub Actions, Argo Workflows, ArgoCD
Deployment Kubernetes: Namespaces, RBAC, Network Policies, HPA
Monitoring Prometheus, Grafana, Evidently AI
Infrastructure Terraform: AWS EKS, S3, IAM, KMS
Security Secrets encryption, RBAC, Network Policies
Testing Pytest for unit, integration, and API tests
Containerization Docker with secure images

💡 Learning Tips for Students

  1. 01-setup → Ensure your environment is ready.
  2. 02-data → Practice data ingestion, preprocessing, and DVC.
  3. 03-src → Implement models, pipelines, and experiment logging.
  4. 04-notebooks → Visualize data and models.
  5. 07-serving → Deploy your first ML model API.
  6. 08-tests → Ensure code quality.
  7. 09-ci-cd → Automate training & deployment.
  8. 10-k8s → Learn secure Kubernetes practices.
  9. 11-kubeflow → Build end-to-end production pipelines.
  10. 12-gitops → Automate deployments via GitOps.
  11. 13-monitoring & 14-monitoring-ml → Monitor infrastructure and models.
  12. 15-feature-store → Manage features effectively.
  13. 16-terraform → Provision cloud infrastructure.
  14. 17-projects → Build real projects for practice & portfolio.

Note: Each folder contains placeholder scripts to be filled in while following the course. Follow the sequence for maximum learning efficiency.