It takes a lot of practice to become good interviewer and majority of ICs especially at small shops never get the required mileage. I don't think i really knew what i was doing until like 100 sessions in...
That's a good part of why I want to stay in the IC track. I don't care about having the power to hire/fire people. That's a manager/lead skill in my eyes.
Maybe we have a small conversation with candidates, but I see no reason I need to be quizing people on their talents.
Every place i've seen where management fully usurped hiring had teams rapidly stacked with extremely poor performers. Not surprisingly so.
Moreover, the days when you as an IC can sit quietly in a corner and churn out code from coffee and jira tickets are rapidly coming to an end. I would highly recommend focusing on those other parts of the job if you're not planning to retire soon, and hiring is part of your job as an SME, misplaced humility notwithstanding.
>Moreover, the days when you as an IC can sit quietly in a corner and churn out code from coffee and jira tickets are rapidly coming to an end.
It was never really like that for me in my domain. But yes, that ended almost 3 years ago for me. I don't think hiring is on my horizon, but I'm looking at long term means to be my own boss.
Just need a few more years first to prepare the jump and pay off debts.
It's almost as if we're saying your resume means NOTHING. I have thought about how to solve this, and my brain comes up with some LinkedIn alternative specifically driven by developers vouching for other developers.
Your resume does mean nothing because anyone can just put anything on it. Sounds like you're basically reinventing reference checks from first principles
Its almost like we have 3-5+ rounds of interviews for some reason.
Meanwhile, I'm sure a 20 minute conversation can weed out 90+% of fake resumes. If the resume was the problem when hiring practices wouldn't be so optimized around trying to find the perfect resume.
1. You still need resume filtering because any public req will receive literally thousands of resumes most of which are poor fit. And that's after HR software did its thing btw. Can't spend 20m on thousands of applications most of which are low effort slop sent by folks who aren't even committed / competent enough to read and follow the req. This was already an issue 10 years ago before AI so I imagine it's at least 10 times worse now.
2. Once you're past that point, at least in my personal experience conducting hundreds of these sessions, most people are pretty bad at going into any kind of depth on stuff listed on their resume so your best strategy (that is if you want to hire anyone at all) is to ask generic questions that are role specific. At best you can tailor some of them to what's in the resume.
1. Of course. I'm not saying a basic filter isn't needed. But if you don't trust your basic filter, how is that going to go when you need to actually invest time interviewing the candidates? You're already off on the wrong foot because your looking for falsehoods instead of qualities.
2. It varies, but I don't think it's a lack of ability so much as liability. My industry is full of NDAs. I can't exactly give precise dates and timelines even for released products. So it's a weird dance of how much I can disclose and how much is a red flag to talk about (since inevitably, this place also probably had me sign an NDA as well).
I'm fine with generic role questions. It usually falls back to leetcode stuff, though.
Thats fair, it definitely sucks for those seeking employment to have a really awful interview. How do you turn it around without looking like an a-hole…
At google where i started interviewing it worked pretty well at least initially when recruiters assembled panels such that you only get 1-2 inexperienced interviewers out of 6. They also didn’t let you do screens as fresh interviewer where it’s much harder to get signal. Every other place i worked it was basically a crapshoot
I get that getting rid of touchid haptic eliminates dead space but still blows my mind they couldn't or refused to figure out screen-based touch id as an option at least. Samsung has it...
Under-screen fingerprint readers are definitely inferior - slower and less reliable. I (Android user) wish they'd revert to back-of-device readers, which were amazing.
(I also wish for smaller screens and no-adhesive battery swaps though, neither of which seems likely to happen.)
I agree with this, while also thinking it was basically physical-enough that the home button still served the same UI purpose about as well as before. But yes it was a step down from the real button.
The haptic sensor is almost as good as the physical button, and the trade off of not having to worry about it breaking (which was likely after a few years with the physical ones) is well worth it for me.
Now with AI bazel maintenance is almost entirely painless experience. I have fewer issues with it than the standard Go toolchain and C++ experience was always quite smooth.
Yes, Captain Obvious. ;) It was before... I can't tell how much before without looking. Looking, it was Autumn 2009.
They included some ex-DANGER people, their architects were big design upfront-oriented rather than anything remotely agile, and it seemed like a stress pressure-cooker. Oh, and a large fraction of product social functionality of those generation of quasi-smartphones were entirely server-side running on a fleet of Motorola's CentOS boxes. They issued workers FireWire-based external backup drives, which I thought was bonkers.
tl;dr: Good riddance to them.
If you have something constructive to add, please feel free to chime in.
reply