Heuristic
Parallelize agents only across separable work
Sixteen agents cannot parallelize one locked door.
When it fits
- You want multiple agents to accelerate a large implementation or investigation.
When to avoid it
- Parallelism can increase cost faster than throughput; compare elapsed-time savings with merge, review and compute overhead.
Why it matters
Find independent tests, modules, research questions or artifacts that can progress without repeatedly editing the same dependency. Assign those in parallel. Keep serial bottlenecks serial until you can split them with an external oracle or clearer interface. More workers on one tightly coupled failure often create duplicate effort and conflicting edits.
An example
Agents can fix independent failing tests in parallel; they should not all rewrite the same shared parser at once without partitioning or interface constraints.
Check your result
Parallel workers have mostly independent failure surfaces and can produce mergeable evidence.
Keep this limit in mind
- Parallelism can increase cost faster than throughput; compare elapsed-time savings with merge, review and compute overhead.
Connected ideas
Use beforeGive parallel agents independent tests and merge boundaries
Evidence and sources
Anthropic's parallel-agent compiler experiment found parallelism useful when agents could work on distinct failing tests or specialized tasks, but wasteful when all agents hit the same serial bottleneck.
The experiment is unusually large and expensive; ordinary projects may not justify many parallel agents.
Building a C compiler with a team of parallel Claudes · Make parallelism easy; Linux-kernel bottleneck