Software Engineer · Phone Screen · Coding

Netflix — Software Engineer ❌ Failed

Level: Senior-Level

Round: Phone Screen · Type: Coding · Difficulty: 6/10 · Duration: 60 min · Interviewer: Unfriendly

Topics: Cache, System Design, LRU Cache, Data Structures

Location: Los Gatos, CA

Interview date: 2025-03-25

Got offer: False

Summary

I had a technical phone screen where I was asked to design an auto-expiring cache similar to LeetCode 2622. The prompt was text-based, and I had to discuss the implementation details with the interviewer. I implemented the APIs, discussed input types, and cache miss scenarios.

Details

The coding question I got was:

`python class AutoExpireCache: def init(self):

def set(self, key, value, timestamp, expire_period):

def get(self, key):

`

My approach:

1. I implemented the basic set and get APIs using a dictionary.
2. I discussed input types and whether to return None instead of False for cache misses (returning None is better because False can be a valid value).
3. I also discussed whether to return an error message and error code upon expiration (which was deemed unnecessary) and whether to allow direct overriding of keys.
4. I wrote basic tests using assert in Python to verify the functionality.

Follow-up 1:

What if memory is insufficient?

I mentioned cron jobs and lazy deletion, as I had seen in commonly asked interview questions. We discussed these approaches, but I wasn't asked to write code for them.

Follow-up 2:

What if memory is still insufficient?

I realized the interviewer wanted me to implement a sized LRU cache. I initially used a dictionary and didn't realize I could use an OrderedDict. Due to time constraints, I didn't complete the implementation.

LeetCode similar: LeetCode 2622