Round 1: Technical Round
Questions: Specific question not provided.
Follow-up Question
- None provided.
Candidate's Approach
No approach provided.
Interviewer's Feedback
No feedback provided.
Round 2: Coding Round
Questions: Given a list of tuples:
[(abc,START,10), (def,START,20), (def,END,30), (abc,END,40)]
<func_name, Start/end, time>
Find the inclusive time and exclusive time for a function like abc.
- Inclusive time is calculated as abc start time - abc end time.
- Exclusive time is calculated as (40-10) - (30-20) for this example.
Candidate's Approach
The candidate solved the problem effectively by recognizing that the exclusive time calculation can be viewed as alternating addition and subtraction. They were able to solve variants of the problem, including cases where abc is repeated and where the start and end of functions are random.
Interviewer's Feedback
No feedback provided.
Round 3: Statistics Round
Questions:
Assume there is a list of numbers which is basically points on the x axis. You have to find the point with the minimum distance traveled from all the points.
Follow-up Question
- None provided.
Candidate's Approach
The candidate initially approached the problem as a one-dimensional k-means problem, proposing to choose a random number as the centroid and optimize from there. However, they struggled to convince the interviewer of this approach. The correct solution involves finding the median of the points for even numbers or any point between the middle two for odd numbers. The candidate took approximately 45 minutes to arrive at this solution.
Interviewer's Feedback
No feedback provided.
Round 4: Data Mining Product Design Round
Questions: The interviewer asked about ways to improve the recommendation system of LinkedIn given new features and limitations, focusing on data and model considerations.
Follow-up Question
- None provided.
Candidate's Approach
The candidate had a productive discussion covering data gathering, cleaning, model selection, training, and validation (both offline and online).
Interviewer's Feedback
No feedback provided.
Round 5: HM Call
Questions: Discussion about the candidate's projects, detailing their contributions and decisions made throughout.
Follow-up Question
- None provided.
Candidate's Approach
No approach provided.
Interviewer's Feedback
No feedback provided.
Round 6: Data Mining Round
Questions: The interviewer, a Staff Scientist, asked comprehensive questions covering various topics including linear regression, logistic regression, decision trees, neural networks, recurrent networks, LSTM, LLM, and offline validations for different model types (classification, ranking, forecasting, NLP, NN, regularization methods).
Follow-up Question
- None provided.
Candidate's Approach
The candidate successfully answered all questions and engaged in a fruitful discussion about the interviewer's projects, providing suggestions on adding a new vertical in LinkedIn search.
Interviewer's Feedback
The interviewer mentioned that the candidate answered well and finished the questions 15 minutes early, indicating a positive impression.