By workload
Cron, but it
actually works.
Leader-elected scheduling. Fires exactly once. Exponential backoff on failure. 90-day immutable audit log.
Explore the details
See your schedule before it starts running.
Choose a recurrence and execution hour. The highlighted dates show a four-week example beginning on a Monday, with all times expressed in UTC.
0 9 * * 1-5 · Times are UTC; no job is created.Example runs across the four-week window.
The calendar illustrates recurrence. It is not a scheduler service or a timezone conversion.
Visual timeline
Every job, every run,
one timeline.
A single leader node owns each trigger. No duplicates across regions or replicas. Retries happen automatically with exponential backoff. The full execution history is immutable.
Leader election ensures exactly one node fires the job
Reliability
Built for jobs that cannot fail.
One fire. Guaranteed.
Leader election across all nodes. Even if three replicas are running, only one fires the job. Verified by consensus.
Smart retries.
Exponential backoff with jitter. Configurable max retries, timeout, and dead-letter handling. No thundering herd.
90-day trail.
Every execution logged immutably. Start time, duration, exit code, stdout, stderr. Queryable via API or dashboard.
Set it. Forget it.
Trust it.
Deploy your first scheduled job in minutes.