Is it possible for you to solve any algorithm question you’ve never seen before in < 15 minutes?

I ask 15 because FB expects 2 problems in 1 interview. How’d you get to that point? I can easily get < 10 minutes on mediums or easies based on ones I’ve seen before or remember, but not new ones. For instance it’s taking me 15 minutes and counting to figure out diameter f a tree (!!!!). What is your thought process like? Do you spend time remembering similar problems or trying to apply known algorithms or ..?

Add a comment
Facebook Seanbean Jun 17, 2018

I solved 3 + a follow up so basically 4 in 45min. I ran interviewer out of questions.

Amazon Grhdsajhg OP Jun 17, 2018

I’ve done that too but I had seen variants of the questions before (and they were easy - 15 lines max). How’d you get so good?

Qualcomm Damn! Jun 17, 2018

You guys are Gods to us mere mortals especially Seanbean doing 4 mediums in 45 minutes. Born to help build social network products. Hail!

Google Rubadub Jun 17, 2018

No

Groupon TC || 👉🏿 Jun 17, 2018

If it's a new question that can be solved optimally using algorithms and datastructures I've studied, then I've done that many times (solved mostly by combining and augmenting them). But designing an entirely new algorithm in a category I've not studied yet, like say something in machine learning or numerical methods or even network flow, is impossible to achieve in ten minutes. The best that can be achieved in that time is a solution. Optimality is impossible since folks who developed those algorithms generally used up a few weeks at least to come up with the original algorithms and data structures for those solutions.

Facebook otwjfpqgea Jun 17, 2018

depends on how hard

Salesforce CmbP63 Jun 18, 2018

I think if you do everything in leetcode, u should see all the questions interviewers ask.

VMware shYC04 Jun 18, 2018

Why bother. Facebook is going down the tubes anyway. The site is crap and finally people are recognizing their dirty business of selling your data. It’s another shitball untrustworthy company

Commvault poll123 Jun 18, 2018

thanks guys for suggestion

Amazon If0nly Jun 18, 2018

15 minutes baseline itself is wrong. I was once interviewing with an SDE3, spent all 45 minutes solving one problem, SDE3 took over the next 45 minutes from next interviewer, we both spent those too on the same problem with the extended problem space I brought up, left with no solution. Still got good feedback and the job.

New
vanierski Jun 18, 2018

What was the problem?

Groupon TC || 👉🏿 Jun 18, 2018

If only, his story wasn't a fiction.

eBay uuop6864 Jun 18, 2018

No

Facebook E7 Jun 18, 2018

You don't always get 2 in one interview. Sometimes it's one 30 min question. Any question can be broken down into common sub problems. These common sub problems are really just algorithmic fundamentals.

Amazon Grhdsajhg OP Jun 18, 2018

Idk, I’m practicing diameter of bst and it seems like all of the solutions are hacks

Facebook E7 Jun 18, 2018

Haha I know the answer and it's not a hack