|
luchadornado posted:Huh? I did 5 PRs today. And it was a lazy day. what were they
|
# ? Aug 4, 2023 19:26 |
|
|
# ? May 28, 2024 00:19 |
|
StumblyWumbly posted:I am also a manager who has often been spread thin, and the squeaky wheel gets the attention[...] That's kind of what I'm realizing. I often do things outside of my role so that the team works smoothly, but then managers tend to leave me alone since I do that. I know I need to go with a facts-based approach into this and ask for what I need, without feeling guilty or that it was my fault, and find a way to stay within my role and ask for help when I need it from other roles. My manager is a relatively newish engineer->manager convert. They have two teams. My team is re-building after a re-org and some departures and is relatively junior - the devs needs a lot of help with prioritization, accountability, etc. The other team is a dumpster fire of bad technology decisions that chew up most of their time. They don't spend a lot of time with my team, largely because I'm helping keep it afloat. Their style scared away 2 seniors on the one team, and one senior on my team. I was able to help stop that trend, but the damage was already done. I really want to go to skip levels and say "they're spread too thin and are a detriment to the team", but that's entirely new to me. I also need them to deal with the unproductive peer first, since that's an even bigger problem. Part of the manager's problem is that this peer has been given a good faith run of about a year and they're actually regressing. I have no idea if they're on a PIP or anything. Honestly, just typing this out and having someone completely unfamiliar is helping me frame how I need to handle this. Rubber ducking rules. Pollyanna posted:what were they Quality of life type stuff: adding a new logging abstraction, adding tests to catch DI/config problems, redoing how a shared library gets published, etc. Trying to help the team focus on the business logic more during a large re-arch. I was just sharing that PRs/day without context isn't terribly useful. A feature PR a day? That's kind of breakneck pace for most teams. luchadornado fucked around with this message at 20:05 on Aug 4, 2023 |
# ? Aug 4, 2023 20:00 |
|
The worst manager I ever had was an engineer trying to convert into management without getting the support they needed to do that without loving it up. It's completely bonkers having a new convert trying to manage two distinct teams while also learning how to be an effective manager.
|
# ? Aug 5, 2023 02:19 |
|
luchadornado posted:That's kind of what I'm realizing. I often do things outside of my role so that the team works smoothly, but then managers tend to leave me alone since I do that. I know I need to go with a facts-based approach into this and ask for what I need, without feeling guilty or that it was my fault, and find a way to stay within my role and ask for help when I need it from other roles. Explicit communication with your manager would help. At the very least make them aware of the problem. Do they already know? I’m kind of assuming not. Also if that person is on a PIP, then you absolutely should have been told. I don’t see how a pip could be done without both tech lead and manager involved.
|
# ? Aug 5, 2023 03:35 |
|
Hadlock posted:Meta hiring freeze officially in effect now Freeze over
|
# ? Aug 7, 2023 22:52 |
|
Lol that was like almost a year.
|
# ? Aug 7, 2023 22:55 |
|
What did their recruiters do during their year off?
|
# ? Aug 7, 2023 23:02 |
|
lifg posted:What did their recruiters do during their year off? Muck up my Linkedin feed begging for jobs.
|
# ? Aug 8, 2023 00:10 |
|
Hadlock posted:Freeze over Meta's reputation afaik is to go really hard on the algorithm lottery. It'll be tough.
|
# ? Aug 8, 2023 01:36 |
|
Yeah they wanted two perfect leetcode medium-hards from me in 45 minutes total.
|
# ? Aug 8, 2023 06:52 |
|
Love Stole the Day posted:Meta's reputation afaik is to go really hard on the algorithm lottery. It'll be tough.
|
# ? Aug 8, 2023 15:18 |
|
Rocko Bonaparte posted:Yeah they wanted two perfect leetcode medium-hards from me in 45 minutes total. Amazing that they think they're hiring the 0.1% of devs who can do that intuitively rather than the 50% of devs who can just memorize it all. At this point they're turning away the 1% who can do even one such question intuitively. I got a surprise question like that in the last 15 minutes of a soft interview with them and couldn't write Java fast enough to complete the question. That was what made me switch to interviewing in python. oliveoil fucked around with this message at 16:25 on Aug 9, 2023 |
# ? Aug 9, 2023 16:22 |
|
I wonder how long it's going to take the tech majors to figure out that their interview format is optimized for a world where the stock market does all their recruiting for them.
|
# ? Aug 9, 2023 16:26 |
|
Tech companies know exactly what they're doing. The ones hustling overtime to memorize Cracking the Coding Interview, looking for the attaboy and a cookie for getting the right answer on the gamified programming challenge, those are perfect worker bees: brilliant and no self respect or boundaries.
|
# ? Aug 9, 2023 16:28 |
|
oliveoil posted:Amazing that they think they're hiring the 0.1% of devs who can do that intuitively rather than the 50% of devs who can just memorize it all. Using Python for pretty much everything that isn’t going to live for more than a couple of months is a lesson I learned early on.
|
# ? Aug 9, 2023 16:31 |
|
Love Stole the Day posted:Meta's reputation afaik is to go really hard on the algorithm lottery. It'll be tough. At the same time, I interviewed with Google for an L5 position. There were five interviews with puzzle coding questions unrelated to the job role. I was told that Google was undecided on my results and that I needed to do two more interviews to see if I was qualified for an L4 position. I declined, since I already had an offer from NVIDIA in hand. At the same time, I tried to interview with Meta for an L5 position. I did not even pass the phone screen, which was a series of theory questions followed by a puzzle coding exercise. After talking through the theory questions just fine, I only had 10 minutes left of the 45 minute interview to do the coding. I did not complete it. The recruiter told me that the phone screen interviewer had not received enough "signal" on my coding ability. I was invited to have another phone screen interview, which I declined. I have had Meta reach out to me about half-a-dozen times since then, inviting me to apply for a variety of roles. Each time I politely decline with "I don't know what you're looking for, but it clearly isn't me. I did not even pass your phone screen the last time I applied. Trust in your screening process, and I wish you the best of luck in finding the right candidate for the role." The Meta phone screen was in C++, by the way. Even using STL, it takes a little bit to write the boilerplate code in there. It was just a blank online text editor where the interviewer said "Write this. Go."
|
# ? Aug 9, 2023 17:14 |
|
hendersa posted:At the same time, I interviewed with Google for an L5 position. There were five interviews with puzzle coding questions unrelated to the job role. I was told that Google was undecided on my results and that I needed to do two more interviews to see if I was qualified for an L4 position. I declined, since I already had an offer from NVIDIA in hand. Didn't you also literally write the book on the topic Google was interviewing you about?
|
# ? Aug 9, 2023 17:34 |
|
Yeah, that sounds like El Goog to me.
|
# ? Aug 9, 2023 17:47 |
|
Vulture Culture posted:Tech companies know exactly what they're doing. The ones hustling overtime to memorize Cracking the Coding Interview, looking for the attaboy and a cookie for getting the right answer on the gamified programming challenge, those are perfect worker bees: brilliant and no self respect or boundaries. hendersa posted:At the same time, I tried to interview with Meta for an L5 position. I did not even pass the phone screen, which was a series of theory questions followed by a puzzle coding exercise. After talking through the theory questions just fine, I only had 10 minutes left of the 45 minute interview to do the coding. I did not complete it. The recruiter told me that the phone screen interviewer had not received enough "signal" on my coding ability. I was invited to have another phone screen interview, which I declined. Ugh, I've been looking at some embedded jobs at Meta because even though the Metaverse is poo poo, the HW for it seems interesting, well funded, and more fun than self driving cars.
|
# ? Aug 9, 2023 17:50 |
|
hendersa posted:I interviewed with NVIDIA for an IC5 (L5) position. There were five straight-forward interviews that evaluated my technical ability with no puzzles. I knocked them out of the park and have been working there for the past 1.5 years. I'm currently being nudged towards an L6 position. What's working at NVIDIA like?
|
# ? Aug 9, 2023 17:51 |
|
Double posting to ask: Anyone have recommendations for advanced books on processor design? Should I just read the ARM specs? I got knocked around in a Microsoft interview because I didn't know enough about how data was cached when going into an interrupt. I swear it was all done in SW last time I looked, but I also had to put chips under UV to erase them.
|
# ? Aug 9, 2023 17:53 |
|
StumblyWumbly posted:This does not match the person I know who recently went to Google. Smart guy, but definitely not brilliant, who likes figuring out brain puzzles, but a chore to work with and no particular talent for actually getting projects done. Their interviews are bad at picking out the candidates they want.
|
# ? Aug 9, 2023 17:56 |
|
Vulture Culture posted:Tech companies know exactly what they're doing. The ones hustling overtime to memorize Cracking the Coding Interview, looking for the attaboy and a cookie for getting the right answer on the gamified programming challenge, those are perfect worker bees: brilliant and no self respect or boundaries. swing and a miss
|
# ? Aug 9, 2023 18:34 |
|
LLSix posted:What's working at NVIDIA like? Based on the described interview process I immediately thought of applying.
|
# ? Aug 9, 2023 18:44 |
|
New Yorp New Yorp posted:Didn't you also literally write the book on the topic Google was interviewing you about? LLSix posted:What's working at NVIDIA like? I had a lot of freedom to automate and investigate things at first, but that has lessened lately with deadlines looming. I get a lot of the "tricky" problems for triage/analysis, which often end up being bugs in other firmware and FPGA components that manifest over on our side. When I present my root cause investigation findings, other engineers are often curious to learn what process I used to troubleshoot because the evidence/symptom trail often wanders all over throughout the system. Most people there have that sort of curiosity. If someone doesn't understand something, they'll find someone that does and ask questions.
|
# ? Aug 9, 2023 18:47 |
|
Sounds really nice.
|
# ? Aug 9, 2023 21:29 |
|
Achmed Jones posted:swing and a miss
|
# ? Aug 9, 2023 21:59 |
|
it was a bad take, no need to double down on it
|
# ? Aug 9, 2023 22:17 |
|
Achmed Jones posted:it was a bad take, no need to double down on it It’s a good take and you’re taking it very personally
|
# ? Aug 10, 2023 00:17 |
|
dont put it in the newspaper that i took it personally!!!
|
# ? Aug 10, 2023 00:29 |
|
Vulture Culture posted:Tech companies know exactly what they're doing. The ones hustling overtime to memorize Cracking the Coding Interview, looking for the attaboy and a cookie for getting the right answer on the gamified programming challenge, those are perfect worker bees: brilliant and no self respect or boundaries. this is a good and correct take
|
# ? Aug 10, 2023 00:33 |
|
Re: interviewing in Python I interviewed at Wealthsimple for a manager position. They asked me to do a relatively simple problem in any language I felt comfortable with, so I did it in Java. Apparently that was the wrong answer since in the next call I had with them I was told that they wanted someone who knew a modern programming language. Maybe I should switch to Python too.
|
# ? Aug 10, 2023 00:43 |
|
the reason it's a dumb take is because you don't actually have to do all that studying crap, you just pull the arm on the slot machine until you hit. it absolutely selects for people who can deal with that- but the "do a 4hr unpaid takehome" thing is 100x worse for selecting people who will do unpaid work, because that's what takehomes actually are. it's trivially easy to get good enough at algo lottery problems to pass within a few tries. it's obnoxiously difficult to get good enough to have a high first-try rate.l (so just don't do that). this means that if you interview at multiple faangs and faang-likes, it's easy (or was, the market is worse now but is bouncing back) to end up with multiple acceptances to play off of each other. there's a lot of poo poo to complain about for those companies, but "it's too hard to get hired" isn't really one of them. however, "the process sucks when you don't hit" is absolutely one. like googles dumbass process wasted hendersa's time, and that's lame. but oh look, they still ended up at a good faanglike place because the odds aren't _actually_ bad when it comes to getting not-terrible questions when taken in aggregate.
|
# ? Aug 10, 2023 00:51 |
|
i guess it does select for people who won't get upset and give up when first-try success isn't guaranteed like of all the reasons to poo poo on these companies that's the only one that makes me roll my eyes. the others (bad labor practices, deleterious effect on society, dumb layoffs, contributing to massive inequality, short sighted products, lol google kills everything, etc etc) are on point tho
|
# ? Aug 10, 2023 00:54 |
|
Achmed Jones posted:i guess it does select for people who won't get upset and give up when first-try success isn't guaranteed Nah the interview process is bad because it has high false positive (gamable) and false negative rates (lottery). It's fine to say it's bad. Because it is.
|
# ? Aug 10, 2023 01:21 |
|
Ensign Expendable posted:Re: interviewing in Python Hell of a leap from "this is one language the candidate felt comfortable with" to "doesn't know any modern language." That's without addressing their definition of "modern" if they aren't applying it to Java.
|
# ? Aug 10, 2023 01:52 |
|
oh yeah, like i disagree re: the amount of false acceptances, and i think that the false rejections are both not really a huge deal and also spread out better than other interview systems that do worse re: disproportionally rejecting pocs and stuff. but i see your reasoning. my thought is p much that it sucks, but it's better than almost everything else i've seen. pair programming on actual features is better, but requires significantly more investment from the interviewer, which makes it hard to scale (and also hard to do for more sensitive stuff if you're not at a bog standard saas shop) my biggest gripe with googles process is that the people who interview you aren't your potential teammates. that sucks so hard on both sides of the table
|
# ? Aug 10, 2023 02:06 |
|
The false positive and false negative rates for a tech major are high, but those are really both side effects of a deeper underlying issue where the interview measures things that are weakly correlated at best with the skills that matter for doing the job. That's less of a problem for junior people, but a serious issue in the L5-L6 range. The interview process also influences who even applies for those jobs in the first place. It's very difficult to say how much that matters but I suspect it's not a total non-issue.
|
# ? Aug 10, 2023 02:17 |
|
yeah i generally agree with all that
|
# ? Aug 10, 2023 02:41 |
|
|
# ? May 28, 2024 00:19 |
|
e: nevermind
biceps crimes fucked around with this message at 02:51 on Aug 10, 2023 |
# ? Aug 10, 2023 02:44 |