The top management have no proper background in ML but think they are doing ML. They are a data science shop with non publication or patent history. They ask questions on regression which is fine but reject candidates if they have forgotten about L1/l2 optimizations. Even principal scientists who are well known in the industry get rejected because of this. I wouldn’t have joined anyway as that would degrade my resume and also end up working with incompetent people no strong theoretical or production backgrounds in real ML.

a recruiter call, followed by a technical screening on data structures and ML fundamentals. The final round includes coding challenges, system design discussions, and ML case studies, often focusing on model deployment and scalability.