Data Science Interview experience of Shaswata Das

Role: MLOps Engineer

1st round:-
Introduction
Current NLP architecture used in my project
How will you identify Data Drift? Once identified how would you automate the handling of Data Drift
Data Pipeline used
Fasttext word embedding vs word2vec
When should we use Tf-IDF and when predictive based word embedding will be advantageous over Tf-IDF
Metrics used to validate our model
In MongoDB write a query to find employee names from a collection
In Python write a program to separate 0s and 1s from an array- (0,1,0,1,1,0,1,0)