Cloud-Based AI Solution Design Patterns

Course details

How do you build scalable AI data architectures in the cloud? What are the best techniques for optimizing distributed AI model training in cloud environments? How can you design resilient and cost-effective AI applications that utilize cloud-based autoscaling and serverless inference? This course delves into proven design patterns tailored for cloud-based AI, addressing challenges in data architecture, model training optimization, and resilient and scalable application deployment and operations. LinkedIn Top Voice and author, Thomas Erl, explains these techniques, and how to use them to create highly effective AI solutions designed to leverage cloud-based services, resources and infrastructure.

Concepts

Introduction

Introduction
What you need to know
Systems and platforms in the cloud
Understanding design patterns

Cloud-Based Data-Centric Design Patterns

Data-centric design patterns overview
Serverless data pipeline
Distributed feature store
Continuous data validation
Hybrid data processing

Cloud-Based Model-Centric Design Patterns

Model-centric design patterns overview
Distributed AI-model training
AI-model operations monitoring and optimization
AI-model drift detection
Federated AI learning

Cloud-Based Application-Centric Design Patterns

Application-centric design patterns overview
AI-workload autoscaling
Containerized AI-model deployment
Serverless AI inference

Conclusion

Next steps