# Engineering capacity with clear accountability.

> Some product roadmaps change every two weeks. In those situations, a fixed-price contract for a year of work can become an exercise in constant scope renegotiation.

Canonical URL: https://marmation.com/engagements/engineering-capacity/

## The decision

Who owns the outcome when the backlog changes?

Embedded contribution needs the client's product and engineering governance. Define responsibilities, standards, access and exit before discussing a capacity commitment.

## The situation behind the request

A product leader owns backlog priorities and architecture, but needs a few additional engineers with relevant AI/product skills for a sustained delivery period.

## Where the obvious solution falls short

Unbounded staff augmentation can create ownership confusion and little institutional learning. It is also easy to confuse an external contributor’s availability with the ability to commit to full product outcomes.

## How we would approach the work

Define roles, capacity, operating cadence, code ownership, standards, decision rights, security access and exit/handover. Pair technical contributors with the client’s product and engineering governance rather than operating a parallel invisible system.

## What the engagement may include

- Named responsibilities and capacity commitments.
- Product/repository onboarding and secure access scope.
- Code review, testing, delivery and reporting standards.
- Knowledge transfer and offboarding expectations.

## The choices worth making explicitly

A fixed implementation milestone is preferable when the outcome and acceptance are already known. Embedded capacity is useful when requirements change legitimately over time.

## What we would test before calling it ready

Measure useful output, quality, rework and communication. Do not sell anonymous seats or promise a team size Marmation cannot staff.

## A sensible first step

Discuss the missing skills, the actual backlog and the accountability required before reserving capacity.

## Related reading

- [Build AI into the product. Not just the prototype.](https://marmation.com/solutions/ai-product-engineering/)
- [Clarity before code. Accountability through delivery.](https://marmation.com/approach/)
- [More generated code is not the same as a better product.](https://marmation.com/solutions/ai-native-engineering-enablement/)