Software Engineer, Product (E4)
Interview Date
March 2025
Result
Rejected
Difficulty
Hard
Rounds
4+ rounds (Initial Screening, 2 Product Architecture, 1 Behavioral, 2 Coding)
Drive Type
Full-Time
Topics asked
Detailed experience
College: Not Specified
Interview Date: March 2025
Interview Type: Full-Time
Result: Rejected
Difficulty: Hard
Rounds: 4+ rounds (Initial Screening, 2 Product Architecture, 1 Behavioral, 2 Coding)
Topics Asked: Data Structures, Algorithms (LeetCode Meta tagged), Product Architecture, System Design, Behavioral
The candidate applied on LinkedIn and got an interview for an E4 Software Engineer, product role.
The initial screening involved two LeetCode Meta tagged questions to be completed in 40 minutes, which the candidate finished.
The recruiter then scheduled a full loop, initially stating it would be 1 Product Architecture, 1 Behavioral, and 2 Coding rounds, but it turned out to be 2 Product Architecture, 1 Behavioral, and 2 Coding rounds.
Behavioral Round: Almost 6 different scenarios were discussed, and the candidate felt the interviewers were satisfied.
Product Architecture Round 1: Involved typical API design for a new user-facing feature in Facebook. This round went "really well."
Product Architecture Round 2: The interviewer was an ML engineer and focused on an infra/system design question rather than product architecture. The candidate started from a product perspective, but the interviewer was "not at all interested in all that" and only wanted to discuss the system. When the candidate mentioned using PostgreSQL for the initial system, the interviewer claimed not to know what PostgreSQL was and asked for an explanation, "condescendingly." The candidate tried to explain it as a relational SQL database and even wrote a SQL query. The interviewer then asked how to improve the query, and when the candidate suggested an index (B-tree), they were asked to explain the data structure and how to calculate the index on every change, including drawing a B-tree and providing an example. The candidate felt "fucked" and that the interviewer was "messing with me," wasting time on index internals rather than the problem at hand.
The candidate received a rejection via email, with no feedback provided due to company policies.