Chapter 7: When Not to Use Agents¶
This is the chapter most books on agents would not include. The most valuable skill an engineer can develop is knowing when not to use a powerful tool.
This chapter is about recognizing the specific conditions under which autonomy adds value, and the much broader set of conditions under which it does not. It is about building the judgment that turns a capable engineer into a trustworthy one.
What this chapter covers¶
- The agent tax -- the real cost of autonomy in latency, money, complexity, and debugging time
- Alternatives that are usually sufficient -- when workflows, rules, retrieval, or static pipelines win
- A decision framework -- a structured approach to the build-or-not decision
- Systems that should not have been agents -- real examples of over-engineering
- Three case studies you should memorize -- patterns that repeat across teams and industries
- Multi-agent anti-patterns -- the most common ways teams add unnecessary coordination
- An honest retrospective on the Document Intelligence Agent -- what actually needed agent autonomy
- Principles for the decision -- six principles that hold across use cases
Code companion¶
This chapter is primarily conceptual, but references the comparison data from earlier chapters:
- Architecture Comparison -- workflow vs agent metrics
- Failure Case Studies -- where agents failed and simpler approaches succeeded
Get the full chapter¶
The complete chapter text is available in the book.