Given a dataset of a billion unique molecule IDs and their descriptions, and a million seemingly random IDs to extract from the dataset, how would you algorithmically architecture the solution in code that you write? You cannot use any databases or tools.
Many questions are asked based on different roles and experience level.
I cannot disclose any questions.
Discussions on the stack you use in your current projects.
Upload database saved in an Avro format to BigQuery.
There were few SQL questions and a question around continuous streaming of data from APIs. The biggest issue was being able to discuss the solution because manager just kept interrupting and didn't provide any hints either. Even before I would complete a sentence, he started writing down solutions even though that wasn't the solution I was suggesting. He continued to move the focus from one aspect to another and then started asking questions about ideas that he assumed I was trying to give even though I repeatedly tried to clarify that what I am suggesting is different from what he is writing down. It was very difficult to communicate and it was also very confusing. In a 30 minute interview I barely spoke for 10 minutes.
Introduce yourself
How to handle severity 1 issues?
Explain Experience with Cross Functional Teams
Data Engineering, Spark, and SQL related questions.
Viewing 1161 - 1170 interview questions