Orchestration Patterns¶
Qin Yu, 21 May 2026, updated 7 Sep 2026
Primarily for platform engineers
These patterns are most directly applicable if you are building or configuring automated agent pipelines — for example, writing a multi-step CI workflow or a research automation script that chains multiple agent calls. If you use Copilot, Claude Code, or similar interactive tools day-to-day, treat this as background reading. The mental models are useful, but you won't wire them up yourself. The exception is Prompt Chaining (pattern 1): the Research → Plan → Implement → Verify discipline applies to any interactive session.
Sub-agents are not just a speed feature. They are a way to control context pollution, parallelism, and failure boundaries. They are also the cache-aware way to use a second model, because a sub-agent's own context leaves the parent's cached prefix intact.
1. Prompt Chaining¶
Use for fixed workflows:
Research → Plan → Implement → Verify
Each stage has a clear input and output contract.
2. Routing¶
Classify the task first, then choose a path:
- cheap model for extraction,
- mid-tier model for routine coding,
- frontier model for ambiguous design.
3. Parallelisation¶
Use for independent work:
- review three modules independently,
- compare multiple implementation options,
- run independent sub-agents for tests, docs, and security.
Parallelism buys wall-clock time, not tokens. Workers fired simultaneously over the same context cannot read each other's cache — an entry is only readable once the first response starts — so N lanes write N entries and read none. Fewer lanes over an identical shared prefix is often cheaper. See Prompt Caching.
4. Orchestrator–Workers¶
Use when decomposition is not known in advance. The main agent plans, delegates, and integrates.
5. Evaluator–Optimizer¶
Use when quality matters more than latency:
Generate → Critique → Revise → Validate
6. Advisor Strategy¶
Run a cheaper executor by default and consult a stronger model only when strategic ambiguity appears. This is one of the highest-leverage patterns for reducing cost without sacrificing quality.