Recommending the right questions for a programmer to solve is a challenge for online education platforms. Nevertheless, this is essential to keep users engaged.
Recommended questions should be ones that are within the skill level of a respective user. The question that this project addresses is "For a given programmer, how much time will they take to solve a given problem?". To answer this question, I built a model to make such predictions based on a programmer's historical platform interactions, their background, and the characteristics of a given problem.
Knowing the expected time for a solution provides insight as to which level of questions should be recommended to a user. The product of this project serves as the basis for a future recommendation engine, which can address a broader question of "What question(s) should be recommended to a programmer to keep them engaged?"