Which setup uses a rules engine alongside a language model?
Why B is the answer
A hybrid flow, with rules for fixed logic
A hybrid flow combines an LLM with a rules engine. Fixed policy suits rules, and judgment suits the model.
In detail
- A hybrid flow uses rules for fixed logic and a model for language.
- Keep hard limits in the rules engine.
Use case
An illustrative scenario, not a real incident.
A loan flow uses rules for fixed limits and a language model to read free text. Fixed logic stays predictable.
See it in different ways
mindmap
root((Foundry agent))
Definition
Role
Goals
Conversation state
Tools
Code interpreter
Custom functions
Tool schemas
Hosting
Prompt agent
Hosted agent
Control
Human approval
Monitoringflowchart TD
Q{"What kind of step?"}
Q -->|"Fixed logic"| A0["Rules engine"]
Q -->|"Language or judgment"| A1["Language model"]Key point: Use rules for fixed logic.
Why the other options are wrong
- A. A second copy of the same system prompt, sent with every single request
- A duplicate prompt adds nothing.
- C. A deployment that has no model
- A language model is needed.
- D. A content filter set to its lowest level
- A filter is not a rules engine.