Describing my previous experience, ML theory (some of which was fundamental stuff which I struggled to recall, but most of it was straightforward and what you'd expect from an ML theory interview), and walking through a case study with the interviewer (presented with a modelling opportunity, asked what things I would need to consider and walking through the steps to get the ML model over the line).
They are very interested in that besides pure machine learning knowledge you also understand the broader business context (i.e. how and where ML can solve business problems for Deliveroo)
How imbalanced data causes issue in Classification ? How is it handled ? What are the evaluation metrics for such scenario ? Which one to choose and why ?
Difference between Lasso and Ridge regression? When to use one over the other?
Specific experience related to the role.
How might customers want to order the search results?
Machine Learning Basic Questions in depth.
ML Theory: Tested theoretical knowledge and core basics of Machine Learning. Topics on logistic regression, different types of loss functions, bias, variance, neural networks, regularisation and its usage.
DSA questions, with sliding window
What is your experience in ML?
Viewing 7481 - 7490 interview questions