Recruitment homework? It depends

If it's uncreative, then yes, it's a waste of time. I've had two that were worth sitting down to.

The most recent one was multitenant in Python, one application serving many clients, each of whom sees only their own data. It was interesting to dig into a topic I hadn't really thought about before. The task got done, but the recruitment didn't work out.

The other task was transport optimisation using Dijkstra's algorithm.

Dijkstra finds the shortest path in a graph whose edges have weights, such as distance or travel time, and it's just right for a recruitment task, because it's a dozen or so lines.

But the shortest path from A to B isn't transport optimisation yet. Things get hard when you have to stop at several places on the way or there are several vehicles. Then it turns into the travelling salesman problem or fleet route planning, and no simple algorithm can guarantee the best result in reasonable time any more.

I didn't get through the recruitment, but I did my homework and got something out of it.