Parallel and batch

Use parallel branches for independent work on one value. Use batch execution for many values through the same pipeline.

Run independent branches

const classifyTopic = new PipelineBuilder(z.string())
  .step((text) => ({
    topic: text.includes('payment') ? 'billing' : 'operations',
  }))
  .build()

const detectUrgency = new PipelineBuilder(z.string())
  .step((text) => ({
    urgent: text.toLowerCase().includes('outage'),
  }))
  .build()

const analyzeTicket = new PipelineBuilder(z.string())
  .parallel({
    classification: classifyTopic,
    urgency: detectUrgency,
  })
  .step(({ classification, urgency }) => ({
    ...classification,
    ...urgency,
  }))
  .build()

Branch keys become output keys. Keep dependent work linear, and avoid parallel writes unless they are idempotent and conflict-safe.

Process a batch

const normalizeTicket = new PipelineBuilder(z.string())
  .step((ticket) => ticket.trim().replace(/\s+/g, ' '))
  .build()

const normalized = await normalizeTicket.batch(
  [
    ' checkout   failed ',
    ' cannot update card ',
    ' password reset issue ',
  ],
  { concurrency: 2 },
)

Batch results preserve input order. Start concurrency low for provider calls and database work, then tune against real rate and connection limits.

batch(...) rejects when any input fails. Return an explicit { ok, value } or { ok, error } object inside the pipeline when the caller needs per-item status.