Factory's Forward Deployed Engineering Model for Droid

Factory describes its deployed engineers' job as helping customers build workflows in which agents can check their own work. In this presentation, the company argues that tests, type checks and other automated validators determine how far its Droid agent can proceed without human intervention.

The distinction becomes concrete in the company's own examples. It describes a constrained internal legal tool as autonomously maintained, while its core terminal interface still needs help with visual problems such as flickering that its validators cannot reliably detect. The engineering task is therefore to create trustworthy feedback for a bounded workflow. Giving an agent a larger assignment does not supply the missing way to verify completion.

“We we like to say this is built, not bought, right?”

Eno Reyes

“The future is here, it's just not evenly distributed.”

Eno Reyes

“If your code base isn't agent ready, you won't see any of the success of the most capable AI systems in the world today, right?”

Eno Reyes

“So if you can frame any problem as the set of verification uh systems that need to validate it, then you can solve that problem with AI today.”

Eno Reyes