SinaraML Definitive Guide

Contents

• Intro
  • Target audience and motivation
  • The problem of Data Science in non-IT companies
  • The most important tasks for AI success?
• MLOps Concepts
  • ML system, ML Pipeline and Model Monitoring
  • Model Serving, Model, Model Service and Model Image
  • ML Product architectures and Model Serving
    • Microservice Architecture
  • ML product dev process and architecture
    • Software dev process. CI/CD and Dev Infrastructure
    • Software dev process vs ML dev process
    • ML product twofold dev process = Software dev process + ML dev process
    • Data Engineering Aspect of Twofold Dev Process
    • Data Architecture of ML product
      • Naive Data Architecture of ML product
      • Complete Data Architecture of ML product
  • From ML Product to MLOps Platform
    • Three ways to harness ML
    • Three Platforms for ML Product development
    • Data Science and Engineering
  • MLOps
• MLOps Platform Concepts
• Data Engineering for ML Concepts
• MLOps Organizational Concepts
• SinaraML Concepts
• MLOps Concepts Advanced
  • Batch Inference Architecture
  • Embedded Model Architecture
  • Real-Time Streaming Architecture
  • OLAP vs OLTP
  • ML product boundary
  • Data Extraction vs Data Integration
  • ML dev process. ML Pipeline and ML infrastructure