Trexquant Investment Quantitative Trader interview experience

It was one of those take-home style coding challenges where the goal was to improve the model’s guessing performance, and the interviewer later asked me directly about how I’d make it better.

After that, I moved into a discussion with an interviewer. The vibe was noticeably machine-learning oriented, and we talked through my Hangman solution and the specific ways to improve it. The topics felt less like generic interview chatter and more like they were probing how I thought about modeling choices, including references to deeper learning and reinforcement-style approaches.

It all felt fairly