Atlassian Interview Experience | SDE2 | Backend

𝗔. 𝗦𝗰𝗲𝗻𝗮𝗿𝗶𝗼 𝗤𝘂𝗲𝘀𝘁𝗶𝗼𝗻𝘀:
- What are the potential issues with using consistent hashing for music streaming servers?
- What are the pros and cons of using pre-loaded hints versus server-loaded hints in an application?
- How would you process a file that is larger than the available RAM on a single system?
- You are tasked with building a sports news classification service that downloads articles and applies machine learning to detect bias. What information would you require to estimate the resources needed for this system?
- When expanding a production-ready application to multiple countries, what backend changes and considerations must be taken into account?

𝗕. 𝗖𝗼𝗱𝗶𝗻𝗴 𝗥𝗼𝘂𝗻𝗱:
- Given a list of words Words = ["baby", "cat", "dada", "dog"] and a random jumbled string like ctay, write a function find(words, word1) that returns the word if it can be formed from the characters of the given string.
- Example:
- find(words, "ctay") → returns "cat"
- find(words, "dad") → returns "-" (not found)

- Given a 2D matrix of characters, determine if a given word exists in the matrix by moving only right or down.

Each file has a collectionId attached. How would you generate a report to show:

1. The total size of all files.
2. The top N collections ranked by total file size.
3. How would you modify the system if multiple collections could be associated with a single file?
4. How would you design and optimise this solution for a multithreaded environment to ensure correctness and efficiency?

- Describe a situation where you successfully handled a project with vague or unclear requirements.
- Explain how you mentored or helped a team member grow technically or professionally.

- How would you design a Rate Limiter?
- How would you scale this system to include a credit-based model, where unused requests are carried over as credits?
- How would you implement and manage this system in a multithreaded environment?

- Design a Web Scraper system that:
- Starts with an initial set of URLs.
- Scrapes all nested URLs recursively.
- Extracts and returns all image links, mapped against their respective parent URLs.
- Follow-up questions:
- Handling depth limits.
- Optimizing scraping for speed and server load.
- Ensuring fault tolerance and retries.