No detailed data science question will be asked they will ask very basic questions and rest is your luck. because you will not be able to know what and why did not like. the question such as how did you remove null value /outlier treatment/ training process.
Questions in Statistics, ML algorithms, Transformers, LLMs, etc. General questions about past projects.
Tell me about previous projects you've worked on?
1. What is Mse and diff between MSE and WMSE. 2.Roc 3 AUC
Explain Transformers Explain BERT Model
SQL and Python pandas database transformation. Walk through a classification problem and caveats (how would you measure performance and feature importance)
1) Mostly on random forest (since it was my project) 2) how would you avoid overfitting 3) What would be your approach if the data is not trainable. 4) Metric 5) Why do you want to join Target
Tell me when you've had the most fun at work
1. A recursion question which has problem statement on no of ways to get promoted, if 2,3,4 promotions are possible per year.
In depth about model performance metrics of supervised machine learning models.
Viewing 3321 - 3330 interview questions