When a job splits into branches, how do you stop one branch from changing another’s data?
- Module
- 02/05
- To read
- ~6 min
Four branches of a document pipeline ran in parallel, and the merge step received results that didn’t agree with each other. What do you think caused it?
In the module you make your guess first, then read what happened and why.
Branch state is scoped by default
Dead-letter routing for unroutable states
Merge nodes validate before committing
Parallel branches must not share mutable references
A plan for keeping branches apart
A spec for your own app, as a text file you download. The module also gives you the prompt for Claude Code, Codex or Antigravity, a checklist to check what it built, and recommended skills.
- Stateless services
- Fault tolerance