Autonomous operations startup StackGen Inc. today launched the Autonomous Operations Factory, a governed layer that lets specialized artificial intelligence agents share context and divide the work of provisioning, deployment and incident response between them.

Underneath it sits Aiden OS, which pairs a shared record of the environment with a harness that enforces policy and logs what every agent does.

AI has fixed only half of the software life cycle, in StackGen’s telling. Coding assistants have pushed up change volume and deploy frequency, while operations absorbed all of it without a comparable step change of its own. The work there is still split across infrastructure, delivery, observability and site reliability engineering, each running on tooling of its own.

Point agents bolted onto each of those functions make the silos faster without connecting them, the company argues. A delivery pipeline can repair a failed build with no knowledge that the error budget is already spent.

StackGen has been collecting numbers on what that costs. Its State of Reliability 2026 report attributes roughly 10% of disclosed outages this year to AI, a six-fold rise over three years, and the company says it has documented at least nine cases since last year in which an agent took destructive action against a live production system on its own.

Aiden OS pairs the harness with the Aiden World Model, a knowledge layer covering what is deployed, what changed, what broke and what fixed it. Agents run against that record under one set of policy checks and approval flows. What one agent learns becomes available to the rest. A reliability investigation opens with deployment history already loaded, and a deployment is gated against live infrastructure drift before it ships.

Four Aiden agents run on it, covering infrastructure operations, DevOps, site reliability engineering and observability. Customers can bring their own agents onto the factory and inherit the same guardrails and audit trail rather than rebuilding governance for each one.

The release also introduces the Autonomy Index, benchmarks delivered inside the product that score how autonomous each stage of the factory is, by team and by application type.

“Writing code with AI is faster than ever; running what it produces is not, and that gap is where enterprises lose money and take on more risk,” said Sachin Aggarwal, co-founder and chief executive of StackGen. “Point agents don’t close it; they just make each silo faster in isolation. A factory needs an operating system underneath it, which is what Aiden OS is: one world model, one harness, one governance model.”

Among the customers already using parts of it is OneTrust LLC. The privacy and AI governance software company automates observability and incident response with StackGen today and is working toward automating the rest of the operations life cycle, its chief product and technology officer, DV Lamba, said.

Enterprise Management Associates Inc. analyst Dan Twing said the approach is sound in principle. Agents working in production need “lifecycle context and governed authority” rather than telemetry alone, he said, and connecting infrastructure, delivery, observability and reliability work through a shared model gives StackGen “a credible foundation for safer, more coordinated autonomous production operations.”

Autonomous Operations Factory is in preview now for Amazon Web Services, Microsoft Azure, Google Cloud and Oracle Cloud users. All four agents are available, and the reliability agent also ships in a free community edition.

The San Francisco company was known as appCD until 2024, when it rebranded alongside a $12.3 million seed round. Backers include Thomvest Ventures, Seligman Ventures and WestWave Capital. Autodesk Inc., Nielsen and Bancolombia S.A. are among its customers.

Image: StackGen