Most Important Skills of Data Scientist?

Don’t give in blindly to demands of ‘infusing’ AI into anything and everything. Companies want to solve tough problems using AI.
But sometimes even simple solutions can solve them effectively. The same goes for deep learning, NLP. Apply transfer learning with caution.

Still not-so-great recruitment practices for Data Science positions.

1. Leetcode problems: Most job-seekers end up forgetting actual data science concepts in their quest to solve these problems before interviews in a
rush (mostly things like inverting a tree etc.)

2. Not having system-design or scenario-based interviews: Companies should ask or pose problems similar to their real-world problems
to see people who would be effective from Day 1

3. Asking cryptic or tricky data science questions to boost ego: Have heard this multiple times when an interviewer asks a question
e.g. How to fill missing values (and refuses to hear case-based answers).

There’s more but unless these are fixed, expect companies to keep crying about not enough ‘talent’ being out there and more attrition.