Uber on-site reject - Where should I improve?
Hello everyone, I had an on-site interview with Uber last week (for Sr.SWE role) and unfortunately I couldn't make it, recruiter got back to me after 3 business days saying it is a reject. I previously posted on tips to prepare (link to that post below) https://us.teamblind.com/s/0m4egEOa I had a really positive experience and I felt all my interviewers were nice & conversational and the whole Uber D-bag culture we keep hearing may not be all true. I want to share my experience and maybe get advice on what costed me this opportunity and where I should focus on improving (not just Uber but any MLLFAAANGU in general). I had 4 rounds of 1 hour each Round 1: Fairly easy round with a Software Engineer (he was possibly L3/L4), asked me an easy coding question on Sudoku that I was able to solve with ease and I was a little surprised this was the only question he had. He was from mobile dev team while I was interviewing for a backend role, could someone from @Uber throw light on whether interviewers could be randomly chosen? Round 2: Hiring Manager round with typical personality and situational questions. "Tell me about a time you failed, biggest challenge you faced, projects would were proud of" etc.etc. I thought this round went good as well although you can never know what the HM evaluated you on. Round 3: Arch/System Design round, interviewer was a Senior Engineer. Talked a lot about his work in his team and scaling challenges we both have faced in our careers so far and talked about designing a fault tolerant CC payment service. I went into detail in some areas where he wasn't expecting me to and at the end I felt I did well (although there was one particular methodology of ACID that didn't come to my head at that time). Overall he looked happy. Round 4: Live coding round where they ask you to code on a laptop, my interviewer was a Senior Engineer. Quite a hard question, I could only come up with a O(n^2) solution, discussed my solution on the white board and unfortunately ran out of time before I could finish the code on my laptop. Since it was past 6pm, my interviewer asked me to email my solution and I promptly did that that very night. I realize not finishing up your code to a working solution is definitely a red flag but I have known people who have been hired despite a bad coding round. Could that one bad round be the sole reason for the reject, although my other rounds were good? Particularly System Design round was pretty good and I increasingly know that is the basis of determining hire/no-hire for experienced candidates. I also would like to put out the disclaimer that I'm an above average Software Engineer who do this to earn a living and by no means a rockstar coder/hacker types who invests all his/her freetime into open source projects and develop an app as a hobby. Thanks for taking the time to read my post and thanks in advance for your response!
Got rejected from Uber too. They didn't provide any feedback. Although I felt the interviews went well, 2 weeks later I realized I made 2 fundamental mistakes. So relax and probably you can do some retrospective to assess your interview performance again.
Thanks for your comment. In the spirit of improvement and helpfulness, could you please let me know what fundamental mistakes you had to correct, if you don't mind?
Most companies don't provide feedback after interview. It's for legal reasons and to avoid liability.
Don’t get bogged down!! I have had numerous experiences like the one you had!! Don’t let this lower your confidence. Take it as a challenge and try harder. Never give up!! 😊 uber lost a good candidate
Thanks for your comment. There is always another chance! :)
Got accept for SE2. Currently, SDE2 at Amazon. Can I ask recruiter to be considered for Sr. SE?
Depends on a lot of factors like availability, your interview performance, YOE etc.
3 yoe. 1yr as SDE2 at Amazon. Do you think I should try?
That sucks. What was your preparation like for über onsite?
Cracking the coding interview, Leetcode and I read this wonderful book "Designing Data Intensive Applications" that gives a massive overview of DS design and practices
That book is huge. Yea just started reading it. Really useful. Thinking of skipping few chapters since I only have 2 weeks.
Rejected by Apple & Uber.. Got offer from Facebook, Google, Netflix, Amazon. Interviews are 50% luck, 50% preparation. You can only control one of it.
You’ve got a great approach to interviewing and it sounds like a shame you could not get an offer. While for legal reasons no feedback is provided in the US for most companies, you could always try to reach out to the recruiter, telling them you liked the interview and if they can share advice on growth areas any of the panel might have seen that you could focus on, should you re-apply after some time. It could have well been the code not running or something not clicking with the architecture person. But with this attitude you’ll get in to a good company.
Thanks @taster. This was one of those opportunities that felt good and everything just went right up until that last round. I certainly have a strong sense of missed out opportunity there.
I'd love to get perspectives from those working specifically at Uber. But from my own experiences onsite at 2 of FANG, my not finishing the working code was the deal breaker, and recruiter mentioned I did well in other areas
Unfortunately, it is usually the deal breaker. If you don’t have the ideal solution, that passes. If code does not work, the feedback often is “not sure how hands on this candidate is...” Companies this size get freaked out from hiring someone who _might_ not be hands on coding.
I completely agree with that from the company's perspective and they have to make a decision on the candidate(s) in the limited amount of time that is as unbiased as possible and since coding round automatically gives a empirical outcome, it is natural for them to decide heavily on that performance. Although personally I have my doubts and I often see coding performance in an interview in limited time is hardly a measure of how someone would do in actual workplace but again, everyone is in the same playing field. So I guess it is a little bit fair that way.
Not being able to finish your coding interview is not a blocker especially on a hard question. I have sat on several such interview debriefs. In general to be hired everyone should be on board. If someone is not convinced and someone else is able to make a case for it you may still get an offer. Conversely if you’ve done well everywhere except with someone who sees a red flag, they can turn the room around. regarding your coding interview. Again, whether you finished or not is not the most relevant. Did you have a sensible initial intuition? If you run out of times you ran into problems. How did you deal? If you were bogged down by something trivial and you haven’t taken necessary steps to avoid it that’s not good (for instance). If you forgot edge cases, or spent too much time on them - not great. If you didn’t test (or found an equivalent way to stay on track) - bad signal. Finally there’s coding style, cleanliness, legibility etc. Lots of things could go wrong (or right) in a coding interview.
These are some great points, I don't fully remember every word I uttered upon hit with this question but I started out by trying to get a brute force solution first and try to optimize it from there. I did talk out loud so the interview always have an idea of what I'm thinking & trying to accomplish. I do remember spending a bit too much time on the white board and ended up with little time to sit back and code on my laptop. As far as testing, I simply ran out of time.
Can you please share what was this hard coding question about?
Was this in LA?
Yes
There is a ton of randomness in interviewing, don’t take anything personally. It could be anything, and improving that one thing might not help you in future interviews. Stay strong and keep practicing.
I certainly agree with the randomness part and I have many skeptical opinions about the whole tech interview process as a whole as it hardly ever reflect the real world work but unfortunately we, as an industry haven't found a better process yet! And no, I'm not taking anything personal, I actually want to have a discussion on where to improve that could be useful for others as well.