I applied for the Machine Learning Engineer at Baidu AI Lab.
A online test as the first round, answer a series of machine learning and Python concept questions and you are given less than 1 minute to write for each question. Then a simple coding task implementing a very basic model in NLP using Python.
A Skype interview as the second round, answer a series of questions regarding typical machine learning models or tasks, most of them are not questions on concept and are practical question that may happen when implementing the algorithms from scratch. Then followed by a very simple coding question on data structure.
What are the causes for gradient descent to converge slow/not converge in different machine learning algorithms?
The following metrics were computed from 8 interview experiences for the Baidu Machine Learning Engineer role in United States.
Baidu's interview process for their Machine Learning Engineer roles in the United States is very selective, failing most engineers who go through it.
Candidates reported having very good feelings for Baidu's Machine Learning Engineer interview process in United States.