Cloudsek SDE-1 Data Acquisition Interview Experience
- Result
- Rejected
- Location
- null
- Experience
- 2 years
- Posted
- 15 Dec 2025
Summary
Applied for SDE-1 position at Cloudsek. Interview process included three rounds with a mix of coding, system design, and Python concepts. Despite thorough preparation, the candidate was rejected.
How it went
Interviewed for the SDE-1 Data Acquisition role at Cloudsek. With 2 years of experience, I went through three rounds of interviews. In the first round, I was asked to rate my Python and SQL skills, followed by a coding question on Roman to Integer. I also discussed Python concepts like *args, **kwargs, decorators, and SQL topics such as query optimization, indexes, and execution flow. In the second round, I tackled two Python problems: extracting patterns between URLs and converting unstructured data into a structured format. The third round focused on system design, including crawling a website and managing concurrent requests. Despite the challenges, I was rejected.
Questions asked (4)
- 01
Roman to Integer
Data Structures & Algorithms · Easy
Open the problemConvert Roman numerals to an integer. The input is a string representing a Roman numeral, and the output is the corresponding integer value.
- 02
Extract Pattern Between URLs
Data Structures & Algorithms · null
Given two URLs, extract the common base URL. For example, from the URLs
http://www.openrce.org/forums/topics/22/12/11andhttp://www.openrce.org/forums/topics/22/13/11, the expected output ishttp://www.openrce.org/forums/topics/22/. - 03
Convert Unstructured Data to Structured Format
Data Structures & Algorithms · null
Parse a string containing name and email information into a list of dictionaries. The input is a list of strings with name, email, and other data, and the output is a structured list of dictionaries with name and email fields.
- 04
Crawl Website and Extract User News Titles
System Design · null
Design a system to crawl a website (e.g., news.ycombinator.com) and extract user news titles. The output should be a dictionary where the key is the username and the value is the news title.