Agentic AI Design Patterns for GenAI and Predictive AI

Course details

Predictive AI and generative AI provide clear business value, but each has its own limitations and risks. By leveraging the autonomous reasoning capabilities of agentic AI, we can now overcome many of those limitations and we can now mitigate many of the risks. In this course, LinkedIn Top Voice and best-selling author Thomas Erl covers a set of AI design patterns dedicated to how agentic AI can be effectively positioned and utilized as part of predictive AI and generative AI solutions. This can result in AI systems that are more autonomous, self-reliant, efficient, and accurate, as well as AI systems that are enhanced with regards to how both humans and applications can interact and integrate with them. This course, intended for AI/ML engineers and AI solution architects, aims to fill in the gaps between agentic AI and predictive/generative AI by showing how these technologies can be effectively used together.

Concepts

Introduction

Introduction
What you need to know
AI agents and AI systems

Knowledge and Reasoning Design Patterns

Knowledge and reasoning design patterns
Autonomous self-RAG
Automated feature discovery and engineering
Causal inference

Orchestration and Creation Design Patterns

Orchestration and creation design patterns
Agent-led parallelization
Hierarchical generation
Voting ensemble

Validation and Adaptive Learning Design Patterns

Validation and adaptive learning design patterns
Escalating evaluation
Autonomous self-correction
Proactive retraining
Shadow model deployment

Conclusion

Next steps