[ OK ] WJUqDqITLJ — full writeup
[ INFO ] category: Behavioral · Onsite difficulty: hard freq: first seen: 2024-05-12
[HARD][ONSITE]OnsiteSDE IIRejected
$ cat problem.md
Company: Microsoft
Position: Data Engineer
Round 1:
- Design a schema for hours spent on application using check-in and check-out timestamps by employee code and a few SQL questions on the same.
- In-depth project discussions
Round 2:
- Longest Increasing Subsequence
- Design an e-commerce site database schema (products, orders, customers tables) and SQL questions on the same.
- Questions revolving around Spark performance optimizations
Round 3:
- Questions on resource configuration in the Spark ecosystem for different sizes of input data
- Design question on data reconciliation & data validation
Company: Microsoft
Position: Senior Software Engineer (Data)
Round 1:
- Heap Sort Implementation
- Merge Sorted Array
Round 2:
- Find Peak Element
- Spark Partitioning, shuffling, broadcasting topics
Round 3:
- Design a data pipeline where you dump data for each day in an S3 storage location, and you need to ingest this incrementally into an Oracle DB.
- How much memory of executors do we need to load 1 TB of data?
- SQL questions
- In-depth project discussions
Verdict: Rejected for both roles as my last round's feedback was not positive.