You hired a coder. You needed an owner.

Every enterprise I talk to this year is hiring developers who are good with coding agents, and building its own systems.

The reasoning is sound as far as it goes. Code got cheap, so hire someone who can now write a lot of it.

It stops one step short. Writing code was rarely where enterprise projects failed. They failed at the requirement, deciding what the operation actually needs, and at adoption, getting people to work differently once the system exists. A developer with an agent makes the typing cheap and leaves both exactly as hard as they were.

Meanwhile the enterprise has quietly taken on every risk a delivery partner used to carry: talent, schedule, outcome, architecture, and the knowledge that walks out when the developer does.

So it is not a staffing decision. It is a risk decision, and it belongs with the CFO as much as the CIO.

The paper also says where building it yourself is right, and which role is actually worth building in-house. It is not the coder: [Xamun Position Paper No. 8](https://www.xamun.ai/whitepaper-the-build-decision), open access.

*First posted on [LinkedIn](https://www.linkedin.com/feed/update/urn:li:activity:7512072626828193792/).*