Module 2 of 7Lesson 2 of 2~17 min
Choose the right workflow pattern
Between a single model call and an autonomous agent lies a whole range of workflows that are more predictable and easier to test. This lesson helps you recognize the main LLM workflow patterns, pick the one that fits your product need and anticipate what it will cost in latency and calls.
Lesson objective
By the end of this lesson, you will be able to recognize the five workflow patterns (chaining, routing, parallelization, orchestrator-workers, evaluator-optimizer), match each one to a product need and estimate its cost.
Topics covered
- LLM workflows
- routing
- orchestrator
- prompt chaining
- AI architecture
Where it fits
Agent, workflow or a single call?
Does my use case justify an agent, and if not, which simpler pattern should I choose?
Lessons in this module
- Decide when an agent is worth its cost
- Choose the right workflow pattern (this lesson)
What you will learn in this module
This lesson is part of the course Design reliable agents and tool calling
- Decide, with explicit criteria, whether a use case calls for a single call, a workflow or an agent, and choose the right pattern.
Related courses
- AI & ProductAdvancedRegular practice: you already use the tools or work on this topic.
Build an AI assistant for your product
3 hr
- AI & ProductAdvancedRegular practice: you already use the tools or work on this topic.
Design a RAG architecture that fits your product
3 hr 30 min
- AI & ProductAdvancedRegular practice: you already use the tools or work on this topic.
Evaluate an AI feature: test sets, metrics and LLM judges
3 hr