"Step function" is sort of a standard term, yeah. Each step is some piece of code that runs at some point in the near future and persists its result for the next step to ingest. It could randomly fail (crashes, preemption, etc) before persisting, in which case the step is retried soon, so it better not have consequential side effects. Most likely you're interacting with external systems (like credit card processors) if you're using step functions in the first place, so idempotency is key.
I've done basically this sorta manually at a smaller scale. Each step was a Postgres table, like "user order initiated," "order confirmed," "order fulfilled," with automated steps (sometimes cronjobs) in between. Sometimes I wanted to expose that status to users. I could see using Inngest at a larger scale.
I've done basically this sorta manually at a smaller scale. Each step was a Postgres table, like "user order initiated," "order confirmed," "order fulfilled," with automated steps (sometimes cronjobs) in between. Sometimes I wanted to expose that status to users. I could see using Inngest at a larger scale.