Testing an idea before building it

The situation

The idea was a multilingual agent, fine-tuned to Australian culture, to help exchange students settle in and stay. Before committing to build it, the university wanted to know whether it would be feasible and worth the cost.

What we did

We ran a feasibility and value assessment over two sessions, using a “paper” prototype to test the idea without writing any code.

What changed

The university decided not to build it. It reached that decision with no development effort and minimal cost.

Why it matters

Knowing when not to build is part of the work. Use AI where it helps, and avoid it where it does not.