Skip to content

sri-gitops/mlops-roadmap-2024

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 

Repository files navigation

MLOps Roadmap 2024

1. Programming

Programming skills are crucial for an MLOps engineer. Python is the most commonly used language in machine learning, making it important for collaboration with machine learning engineers and data scientists.

1.1. Python & IDEs

Start learning Python through books and practice.

1.2. Bash Basics & Command Line Editors

Understanding bash is essential for MLOps tasks.

2. Containerization and Kubernetes

These technologies are vital in modern software engineering.

2.1. Docker

Docker is a popular open-source containerization platform used in MLOps.

2.2. Kubernetes

Kubernetes is essential for machine learning model training and deployment.

3. Machine Learning Fundamentals

An MLOps engineer should have a basic understanding of machine learning models.

  • Courses: MLCourse.ai, Fast.ai
  • Book Suggestion: Applied Machine Learning and AI for Engineers by Jeff Prosise

4. MLOps Principles

Awareness of MLOps principles and maturity factors is required.

  • Books:
    • Designing Machine Learning Systems by Chip Huyen
    • Introducing MLOps by Mark Treveil and Dataiku
  • Assessment: MLOps maturity assessment
  • Great resource on MLOps: ml-ops.org

5. MLOps Components

MLOps platforms consist of various components, from version control to feature stores.

5.1. Version Control & CI/CD Pipelines

Critical for traceable and reproducible ML model deployments.

5.2. Orchestration

Systems like Airflow and Mage are important in ML engineering.

5.3. Experiment Tracking and Model Registries

Logging metadata, parameters, and artifacts of training runs.

5.4. Data Lineage and Feature Stores

Feature stores are a crucial component of MLOps infrastructure.

5.5. Model Training & Serving

Decisions depend on the organization's infrastructure.

5.6. Monitoring & Observability

Vital for the health and performance of ML systems.

6. Infrastructure as Code: Terraform

Essential for a reproducible MLOps framework.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published