Engineering Beyond the Model

The model is rarely the whole system.

A useful AI product also depends on the quality of its knowledge, the boundaries around its actions, the way people verify its output and the architecture that allows every part to change independently.

This collection documents how I think about those systems.

Working principles

1. Architecture should outlive today's model.
2. Organisational knowledge is more valuable than a clever prompt.
3. Privacy is an architectural decision, not a checkbox.
4. Automation should improve judgement without removing accountability.
5. The right model depends on the task, cost, latency and risk.
6. AI should fit into the organisation's workflow, not force the organisation to fit the AI.

What changed my mind

I once thought the largest challenge would be choosing the most capable model. Building real systems made it clear that the harder problems sit around the model: context, ownership, evaluation, permissions, trust and long-term maintainability.