Our test is a shortish take home exercise: implement a simplified version our core business solution (a matching engine). I judge candidates on: - choice of a

Our test is a shortish take home exercise: implement a simplified version our core business solution (a matching engine). I judge candidates on: - choice of abstractions - quality of the tests - readability (very important, reading code is x10 harder than writing it) - knowledge of the language and it's libraries - correctness of the solution I penalize candidates for: - reinventing the wheel (I.e. Roll their own sorting algo) - overengineering - unnecessarily convoluted, 'clever' solutions - premature optimisation