Explains roadmap dependency and decommissioning order
Dependencies point to the same or an earlier state without cycles; an application is switched off after its consumers move.
Can AI do this?
AI does this well Most of the tested AI models master this. 3 of 4 models in the current benchmark master it.
| Model | Tier | Probability of mastery | Status |
|---|---|---|---|
| gemini-4-argon | Flagship | 100% | Mastered |
| gemini-3.5-flash-lite | Low cost | 96% | Not shown |
| gpt-6-luna | Low cost | 100% | Mastered |
| claude-haiku-5-5 | Low cost | 100% | Mastered |
Mastered means a probability of at least 90% that the model has this knobit, estimated from its graded answers. “Not shown” means the answers do not settle it either way. How the models are tested.
What it looks like in practice
Expected result: Dependencies point to the same or an earlier state without cycles; an application is switched off after its consumers move.
How it is assessed: Answers a written question on the roadmap sequencing correctly
Related knobits
Builds on
Needed for
- A-501 Recognises a forward dependency in a roadmap
- A-505 Recognises decommissioning before the consumers move
- A-506 Recognises parallel changes to one system of record
- A-511 Recognises a dependency on an unknown package
- A-531 Resequences decommissioning after the consumers move
All 14 knobits of Sequence the implementation roadmap.
Source: Practice the stage 5 reference and review checklist rest on