Case studies / StrongDM

A three-person AI team and non-interactive development

StrongDM’s AI team published a first-party account of a “Software Factory” in which humans write specifications and scenarios, and agents write and iterate on code without human-authored patches or human code review. The validation story is scenario satisfaction against a Digital Twin Universe (behavioral clones of Okta, Jira, Slack, and Google Workspace APIs), not a conventional green test suite.

Dated 06 Feb 2026

Scale claims

• Verified The StrongDM AI team was founded on 14 July 2025 by Justin McCarthy, Jay Taylor, and Navan Chauhan.

  factory.strongdm.ai/ — Named individuals and date are on StrongDM’s own factory site. This is a team-formation fact, not a throughput metric.

• Verified The factory’s operating rules are that code must not be written by humans and must not be reviewed by humans.

  factory.strongdm.ai/ — Stated as rules on factory.strongdm.ai. We are not independently confirming that production StrongDM product code follows those rules.

• Verified The Digital Twin Universe can run thousands of scenarios per hour.

  factory.strongdm.ai/ — First-party capability claim (“thousands of scenarios per hour”). No method, no measurement window. Treat as an order-of-magnitude statement from the builder, not a benchmark.

• Dropped / flagged A working factory requires at least $1,000 per human engineer per day in token spend.

  factory.strongdm.ai/ — Published as a koan / “practical form” on the same page, not as a measured production figure. Dropped as a metric.

Sources

• https://factory.strongdm.ai/
• https://www.strongdm.com/blog/the-strongdm-software-factory-building-software-with-ai

Named tools

Only products the primary source names, and only if we list them. cursor