Companies are still debating exactly how software development in the area of AI should work, but one of the first answers is the so-called software factory. Essentially an agent cycle built around the traditional stages of software development, the software factory approach has become a popular way for companies to remake their engineering organizations for the age of AI.
Now, a Warp system could make that transition much easier. On Tuesday, the AI coding company unveiled Deformation Factoriesa new system designed to make building and running AI software factories as easy as possible.
Serving as an infrastructure layer, Warp Factories provides companies with a simple environment for deploying agents and a roadmap for how to use them.
To be clear, many companies are already finding success with the factory model without the help of Warp. Stripe has been particularly public about its technical progress, developing a system of “subjects” to automate development within your own code base. The ramp has made similar progressby developing a background agent that can monitor its own code after it is deployed.
As Warp CEO Zach Lloyd sees it, Warp Factories’ target market will be smaller companies without the resources to develop a system from scratch.
“[If you look at] things like running your agents in the cloud and directing them as they run, or bringing the work they’re doing to your local environment, or configuring the memory that goes through those agents, or configuring assessments that goes through those agents; It’s actually a huge infrastructure task to get this right,” Lloyd told TechCrunch.
At Warp Factories, the architecture is already built out of the box and many of the most difficult decisions have already been made. Warp’s system is based on the standard phases of software development (classification, specification, implementation, review and verification), but the agent approach means that any of those steps can be automated.
Users can choose their own coding model and harnesses as needed; The system works as well with Codex as it does with Claude Code. It also integrates with ticketing systems like Linear and Jira, and messaging systems like Slack and Teams, in an effort to seamlessly connect to existing workflows.
Beyond shipping code, Warp Factories will also give managers the tools to track factory performance. With all agents running in the same environment, it is easy to compare performance metrics for different configurations and monitor overall token spending. Warp Factory also allows for self-improvement loops to optimize the overall system, automating the management of the process itself.
Still, Warp Factories isn’t designed to completely replace software engineers, just to give them an easier way to collaborate with the new agent workforce. In Lloyd’s own experience, there are still many tasks that require a human being behind the wheel.
“We automate about 30% of our tasks, 30 to 35% on a weekly basis,” Lloyd told TechCrunch, “and as the models, context and harness improve, I think that number will increase over time.”
When you purchase through links in our articles, we may earn a small commission. This does not affect our editorial independence.