I had a long conversation with Claude.ai. It was mostly about interview techniques for hiring software engineers, but as is my habit, it got philosophical.
At one point I complained about timed programming tests. One problem I have with then is that it pushes you to code so you don't run out of time. So the first idea you have on the solution approach is generally the one you stick with. But a good engineer will consider multiple approaches. A *really* good engineer might be dissatisfied with their current solution set and bounce ideas off a collaborator to get fresh thinking. A timed test pretty much precludes any of this and only measures if the best solution pops into their mind within the first 30 seconds. That's not what I hire for.
At another point I mentioned that at one point in my F2F interview, I ask for two solutions to a problem that we investigate together. I expressed surprise at the number of candidates who can only think of one solution, and keep trying to dress it up in different clothes as their second.
Claude crystalized those two observations with "Hold our own ideas loosely." This really resonated with me strongly for several reasons:
- Avoid "not invented here". We want good ideas no matter where they came from. Leave your ego at the door.
- Closely related: Avoid attachment. Once you've invested in your idea, don't hold it too stubbornly, especially in the face of new information. Don't fall victim to the sunk cost fallacy.
- Don't let your first idea blind you to further thinking. Re-frame the problem in different ways. (In a previous post I described falling victim to this in ... wait for it ... a timed coding test.)
- Embrace collaboration. Two minds will usually produce a better solution than either one in isolation. Some of my most rewarding work experiences came from working with another engineer where our final solution was a wonderful synthesis of our two starting points. I find that exhilarating, not demeaning.
