$830
for a full back office — $4,211 uncached
Fifty-two instances, one orchestrator, two findings caught pre-ship
One orchestrator ran a 52-instance fleet to deliver a 10-phase back office for a multi-tenant platform — 197/197 tests, 24 migrations, 26 commits — in 7h14m wall clock. Four of the agents were reviewers; they caught two blocking security findings before the campaign could close.
51 subagents across 11 workflow runs plus 5 direct spawns, ~72 minutes of orchestrator active time, 48 tasks completed. The fleet isn’t just throughput — four of those agents were pointed back at the rest as reviewers, and they caught 2 BLOCKING security findings that were fixed in-session before close.
Compute cost was $830.24 actual against $4,211 uncached. The agency-rate equivalent for the delivered scope is roughly $10–20k — the same work, one orchestrator, a fraction of the spend, with the security review built into the run instead of bolted on after.
- Part of the fleet reviews the rest — security findings caught before ship
- 197/197 tests, 24 migrations, 26 commits in one campaign
- $830 actual vs ~$10–20k agency-rate for the same delivered scope
From the session record
One orchestrator, 51 subagents across 11 workflow runs plus 5 direct spawns: a 10-phase back-office build for a multi-tenant platform, completed in 7h14m wall clock with roughly 72 minutes of orchestrator active time. 48 tasks completed, 26 commits, 24 migrations, 197/197 tests.
Four of those agents were reviewers, not builders. They caught 2 BLOCKING security findings that were fixed in-session, before the campaign was allowed to close. The fleet isn't just throughput — part of it is pointed back at the rest.
Fleet metrics from a campaign on a dance studio management SaaS (client work).
This is what Pro delivers.
Not features for their own sake — measurable leverage on every session.