- Published on
Meta | SWE (E4/E5) | NYC/Remote | 2024-01-19 [Pending]
- Author
- Shared Anonymously
Status: Experienced, Senior SWE, Employed, 10+ YoE, State school BS EE,
Position: E5 SWE @ Meta
Location: NYC
Date: January 18 024
Technical phone screen (1 hour):
- Two standard common Meta tagged questions. Leetcode Medium to Easy. Speed is key as you must complete both problems with optimal or near-optimal solutions to proceed.
Virtual on-site
-
Standard behaviorual interview. I prepped using the guide on Meta careers and had written down detailed bullet points for the prompts on the Meta career website. Unfortunately, not a single one was asked and instead I got the usual "tell me about a conflict, tell me about a tight deadline" that I should have expected but neglected to prepare deeply for. It doesn't help that my background is in finance so the traits that Meta looks for are sometimes diametrically opposed to what I have been doing (slow and very careful for a long time vs the good enough with brute speed that Meta wants). The problems I talked about also didn't resonate with the hiring manager. In hindsight, I should made some plausable sounding and relevant drivel that woudl resonate better.
-
Two coding rounds I Just like the phone screen, two questions each round, with 40 minutes or less to do both. All the problems were all tagged Meta and are very common. Unfortunately, in my rush, I tangled myself by overcomplicated one problem despite having an optimal verbal solution and writing the algorithm in English. I did mention how I would fix it if I had a bit more time. For two other problems, I wrote correct, optimal code (or so I thought) but perhaps slightly more complicated than what I would have liked. I also used some newer Python syntax, etc. For instance, I referred the interiewer to the python reference when I said the heapq API doesn't accept a custom comparator (mostly true) which is why I had to add some roundabout code to handle a different comparison criteria.
-
System design- I will add more detail when I update this post with final results but it was not one of the questions on this list but it was one of the dozen or so common questions not on that list. I spent a lot of time discussing how to handle very high concurrency and contention for a feature that seemingly required it. The interviewer was looking for a specific approach that mitigated this and wanted the exact technique which escaped my mind. I think this interview has the fatal mistake that will result in a rejection.