it is insane, been on like three figures hiring squads at this point and probably half the people are total bs dead on sight, half of what is left actually just played buzzword bingo with HR and got in. maybe like 2-3 in 40 are useful to interview but maybe you interview around 10 with some of the not sure people in there from those there’s always now people using ai on the fly, super obvious and always disappointing, then some just have a scripted reality and those area real pain to disambiguate because they talk the talk real well - but you put them in front of a problem and they break. the most painful are the ones who talk the talk, can mostly walk the walk, but are so opposite to everything the position is, in sec/ops see so many people who actually just want to do dev but they also know a little k8s and are attracted to the pay another common one is people who don’t really like what they doing and would rather be the manager of that thing, so they really push on climbing higher, usually in front of the guy they would be replacing, but like wrong place, wrong time, wrong guy by the end of a recruiting round you usually whittled it down to a couple who potentially fit, then gotta roll the dice. inevitably that then results in a group decision despite obvious (to me at least) objections- the person ends up joining and being useless and untrainable and now thanks to hiring/firing laws really hard to shake (if they last through probation, which is basically a social game more than technical aptitude) the amount of hiring rounds that start with liars and frauds, then end with the accepting of a good liar and fraud, eventual dismissal- and repeat… genuinely baffling how anywhere can blind hire. best hires ive ever made were genuine, interesting and interested, knew their shit well, socially adapted, and hit the ground running, recommendations are a huge resource for this kind of person in a lot of cases (assuming the rec is coming from this type of person, naturally) the feeling of that person completely unique but to the corpo side they can’t seem to sense the vibe and just take the person who can yap the most convincingly. it is the liar’s hunger games hands down, whoever can yap the longest lasts the longest, may the best bs spewer win. really hard to fight against, usually a case of stacking “i told you so” moments so people believe in the mystic power of your bullshit detector and vibe-o-meter, eventually you become the one who has to help filter the top level and that’s when you really see the crap people come in with. wild world we live in, genuinely insane and infuriating problem thanks for coming to my YAP talk.
honestly the best advice is to know your shit impostor syndrome is mostly just an internal monologue actively devaluing you, largely due to comparison to external things nobody is saying be best of the best, but if you are applying to a kubernetes role then know that, again not a perfect expert (unless you going into some kind of principal/specialist level role then the bar a hell of a lot higher by default) you want to know all of the “basics” as if it is breathing, terminology is important (you can SRS those for a cheap win), keep up to date with the new developments and read the release notes you’re going to seem 100x more interested if you are actually interested, so those things are high signal - if you are asked about things you’re most excited about in k8s (to stick with the example) and you just roll of the dome that: “cross-core cpu distribution seems like it is going to change things up” then suddenly you’re like top 1% of shit-knowers in their eyes and might know more than the guy hiring (because corpo always a bit behind and only care about their next version usually) next thing: practice, always be doing stuff, do the stuff that you want to be good at, if it is k8s then run KinD, run a mini homelab, break 20 k3s instances, enjoy the process, do things for the sake of just doing them, if you have something to show then great, if you have 10 things you did that you can talk through then perfect don’t be afraid to not know something super complex in the moment, chances are if you do the above then you can probably intuit it and fill in the gaps (think aloud here, ground truths and build on it, don’t talk nonsense), often there’s like 50% “testing knowledge” and 50% “can this guy reason about a problem” i’m never afraid to engage in a dialogue when someone doesn’t have perfect knowledge about exactly what i’m asking and they tell me what they do know about that some of the better problem solvers i’ve seen will think through a problem out-loud, reinforce it with their axioms, relate that back to the question, speak about a theoretical way through the problem (usually while mentioning that it only works in the case that they have the right idea), we stop and discuss, we iterate, it becomes a group problem, there is no spotlight anymore and now we know this guy has the right stuff going on, they’ll figure it out given their already great baseline. super important and high signal trait the corpo/HR (non-technical) sections, just radiate, best foot forwards and sell yourself well, know what the place you’re interviewing at is and does, know their mission and their “why”, keep it kind of casual-professional, you don’t want to be a suit but you don’t want to be the other extreme either, hit it about 3/4 of the way to suit for real though, if you’re going for a role and you want that shit locked down, then lock in, go deep, go wide, let the whole role wash over you at a minimum, for what you specifically want to do that’s your soul, be unreasonably good at it overall the fundamentals need to be trivial, esoteric stuff can be really cool, if you accidentally teach your interviewer a thing and you end up in a dialogue driven by mutual curiosity then they are gonna sing your praises in the post-interview meeting it isn’t about knowing everything about the thing, it is about knowing your shit, be unshakeable in what you know about what you do, you want to have solved so many problems that hypothetical scenarios are like water off a duck’s back if you want people to believe in you, you have to believe in yourself. also go apply to a bunch of stuff even if you don’t want it, learn the game, get comfortable, nothing should be a surprise, you don’t lose anything for taking 20 interviews another super long one, apologies for the wait, just didn’t want to give a few hazy bullet points and leave you with 10 more questions spaced repetition system (one of the most common being software like anki)
algorithmic information retention maximization, insanely powerful though basically brute force, language learning community basically runs off of it (and medical/legal/scientific etc. study)
most important thing will always be doing things, have a wealth of projects that dig into a variety of problems, try to keep them small and doable in a few hours up to a weekend (ideally they shouldn’t need planning more than a rough outline and first point of attack perhaps) it boils down to: quickly building things you need or are interested in, because you can, solve a problem, also no reason you can’t solve a problem that is already solved - who is gonna stop you figure out how ANSI sequences work and build a simple tui from scratch if you figure out http you can build a http server from the ground up if you have any hardware like a stream deck or a gamepad then learn how to take those raw signals and translate them into something happening in an app make a websocket sidecar that gives you a nice website based gui for your local dotfiles (and think about why that might be more useful than an electron app with system write access) hit some APIs and see what they give you, make a cool dashboard, automate something you do regularly, glam it up just for show make sure to use common marketable and hiring languages, zig is really excellent but I can count the jobs going on one hand, things like c#, java, python, javascript, php - those things are what the market demands (you can always play with new langs in your free time once you’re employed - like me!) most important things for selling yourself technically in any kind of interview situation will always boil down to:
- know your shit, fundamentals should be trivial, esoteric stuff very cool but might not come up
- what have you done that you can talk through the process of, what did you learn doing it, what would you do differently next time, what are some of the challenges you hit those are all super common questions in interviews if you have 10 things to speak about that are all cool deep dives into a problem that interested you, done in you spare time - and the other new grad has one carbon copy project they did because they had to guess who I’m more likely to back