Job Interview Tests

[@MAYOFORCE]

You got me here pausing and answering each question as you ask them lol

[@MisterShine7]

I had a lot of fun pausing and thinking about my answers to each of these too

[@wabba67]

Same. I've never even applied to positions in the games industry but this was fun and something different.

[@Letterface]

"Given two integers, return their average." Oh you sneaky bastard. Now that's an int score

[@yourwifesboyfriend6907]

Ok but did you come up with anything worthwhile besides the obvious solution?

[@Letterface]

@@yourwifesboyfriend6907 If I had something worthwhile I'd have the job instead of frustration with the difficulty of needing to be a programmer before I'm allowed to be a writer

[@stuartmorley6894]

This reminded me of an interview for a job I got. They questioned me about a bunch of stuff I had done. They then asked about using a particular piece of kit and the experience I had of it. I answered honestly that I knew what it was but had never actually used it. I found out later that they'd given me the job in part because I'd been honest about what I couldn't do. I'm not sure how often that would have happened at other jobs but it made me think more about trying to bluff my way in the door.

[@courier665]

In my field job interviews boil down to 1 question "Can you pass a drug test" And 7 times out of 10 they can't and still get hired

[@talkingmudcrab718]

11:07 Job Applicant: "My favorite genre are games with lots of microtransactions..." **Tim Cain rips a mask off of the job applicant** Leonard Boyarsky: "Hey! It was Old Man Bobby Kotick this entire time!" Old Man Bobby Kotick: "Yeah! And I would've gotten away with it too if it weren't for you pesky kids!"

[@DanielTralamazza]

I used to do exactly these "implement atoi()/reverse string" Google interview kinda questions, but with time I experienced a few issues: 1) People study for them (basically Goodhart's law). 2) I am not hiring people to work under pressure of losing their jobs (essentially what an interview is). 3) As you alluded, work is not slide pizza in code comes out, same for impromptu questions. I am looking for motivated and self-reliant people, so I eventually changed to this: I ask them to show me their best code (100+ lines), one they are proud of, and I ask them to explain me exactly why. This checks many boxes for me: 1) The obvious ones: coding style, clarity, correctness etc 2) What *they* consider good code and *why* 3) It's instantly clear if the code is copied or studied, only an author can explain the minutia At the very end I ask what they would have done differently.

[@mussdem]

We need more interviewers like you, I am shit at those kinds of tests because generally the stress of the interview makes me make dumb mistakes that while on the job I could take a break and come back and see it with a fresh set of un-stressed eyes, whereas in an interview you can do no such thing. Talking about code I've written has always landed me a job, trying to code some crap leetcode question always landed me a thanks for trying email. I really hope the code tests in interviews go away entirely, they are completely unproductive and will lose you engineers that are good at the job but bad at the interview as opposed to engineers that are bad at the job but good at the interview.

[@Dustb0]

I actually really like this approach from both sides.

[@SPTX.]

@@mussdem holy shit bro, this so much. The amount of people I have seen being a liability at work being there only because they aced the interview is insane.

[@jawharz9759]

I like this a lot since I really hate those algorithm tests but idk how I could show code I wrote while working for another company 😅

[@roberteltze4850]

I like your approach but I've never once taken code out of the company I was working for. I could discuss my best code but I couldn't show it to you.

[@austinbevis4266]

I’m going to apply to work for Tim under multiple aliases so I can see all the hidden dialogue options and speech/skill checks.

[@Geophaser]

I've conducted a lot of code & design interviews and Tim is 100% right here. The number 1 thing I'm looking for in an interview is how someone reacts to making a mistake or doing something un-optimally. If there's a lot of "oh that's what I meant" or boasting without substance it's a massive red flag. So many people, even (maybe even particularly) very experienced designers and programmers, basically try to bluff their way through interviews.

[@jeffreyedwards9968]

Similar to my approach. I've done a handful of interviews and my "big question" is asking for a "war story", some specific time something went badly and how it was handled. I think it's useful to find out how someone identified and responded to (not even necessarily fix) a real problem with the intent that I understand it. The answer should be unique to their experience and should demonstrate they have been through this rodeo enough times to speak frankly and specifically about the work. And who wouldn't be proud to talk about turning lead into gold?

[@UENShanix]

Completely agree from the same position. My favorite combo ice-breaker/experience evaluation questions are "You go back to the start of the last big project you worked on, what's the one thing you think you nailed right out of the gate," followed by "And what's the one thing you wish you could've gone in a different direction with?" They're introspective professional questions and allows someone to puff themselves up a bit, which helps people feel at ease and tells me what they really liked about their last position. Helps me identify their strengths. And the second question is great for the real introspective part because, ideally, a candidate responds with what was "bad", why it was "bad", and what they've learned from it. It's easy to bluff the first question but you can tell if you're being fed some grox if you get told that everything was peachy keen the whole way through.

[@adamsbja]

Decades ago I made it to the "bonus round" of a middle school math competition, where after the main competition the top scorers are given a problem and then need to give a short lecture about their answer. It was something about reflections of magic squares. Everyone else was from the same school where as part of their drilling they happened to have had "there are this many magic squares, here's how you know that" so they could start from there and go on into the meat of the question. I didn't know that, so I spent my timeslot trying to figure that part out, running into dead ends and talking through what I was trying to do. Didn't win any awards, but a couple of the judges told me afterward that they were really impressed how I handled not knowing the answer and not knowing how to get there. Looking back now that feels pretty dang good.

[@TheAlison1456]

what's wrong with saying "that's what I meant"????

[@Geophaser]

If someone says one thing that is kind of wrong or adjacent to the answer, then you give them the answer and their response is "that's what I meant", it's as if they are trying to retroactively get credit for something they didn't articulate, rather than trying to learn from the experience. @@TheAlison1456

[@94Pattycake]

As someone else mentioned in the comments, funnily this video has given me more "confidence", that would not be the word I would use but I feel more grounded. Thank you all of the giving back you do.

[@jsj0520]

This entire video makes me feel even better as an interviewee. I felt comfortable and confident answering each of these design questions. The only problem is trying to land an interview in the first place, out of the hundreds of applications I’ve sent in the past 2 years since graduating college I’ve only ever had 1. It was for Rockstar as a “world event designer” but I never played any rockstar games and I don’t find them that entertaining. (I only applied because I was looking for anything and my parents say a “shotgun” approach for sending out applications is best which I completely disagree with)

[@JasonStorey]

One of my favourite personal life lessons was actually very analogous to the playlist question. I wrote an android based barcode scanner app. It was used on dedicated trigger "gun" scanner devices. I designed my ui to be very user friendly and pretty. It included a modal dialog of success on correct scan api response (you were scanning and connecting a product to a label) turns out it was a touch screen button and forced the users to take their hand off the product to tap the screen each time. they hated it. I removed it and it made their work 40% faster. I now think a lot more carefully about seemingly small decisions like that. I aim for the "rule of expectation" and side-eye any blocking ux option. in the playlist case here for example, I'd popup a notification the first time you do it with a dismiss button and a button to change the global rule. If you click another song in the meantime it would auto dismiss. In short, The expectation was to add a song. I'd add a song. There is a world of difference between a mistake and more context. I would aim to passively give them the context needed to decide if they _want_ to do this, but I would do... what they expected would happen. Alternatively I would let them make the entire playlist and have a small label saying "X duplicates" with a single prune button. I do like questions like that though. no right answer outside of "validated learning" from users, but its always a fun thought exercise and a great way to remind you what the point of ux design is.

[@deathsheadknight2137]

I wish more people put thought like that into their designs. I've seen some software where they added one, two, three or even four extra superfluous button presses that they could have simply automated, streamlining the entire thing.

[@veranish]

I've had almost all of these design and code questions asked of me as a technical designer (and 100% of my interviews have been remote since covid), thanks a ton for being this specific and direct! This is an amazing addition to your how to get hired video. Re: take home tests, it does seem to be the industry norm for entry level, I have not yet had an interviewing experience where a take home wasn't expected (often in addition to live coding or design questions), I've done maybe five tests now (some were really easy to the point that I don't call them tests, others were a week long and really arduous). I have three follow up questions mostly from the perspective of an interviewer though I'm still an interviewee these days: 1: What are good questions to ask yourself if you think you might be taking "People lie" too far? I've had some negative interview experiences where a semantics difference led to hostility in the interview that I had lied on my resume, (one specifically over the definition of what a "tool" is or isn't, seemed like an over zealous attempt to catch me in a lie when really we just had different definitions of in-editor tools being a "tool" vs only out). 2: Do you believe it's important for interviewers to also let the candidate know how you work (assuming they'll be working together) so they can equally judge if working there will work for them as much the same as you, and how would you facilitate that? and 3: would you have anything more to add about interviewing entry level vs mid vs senior and how you might verify their experiences differently? Thanks a ton for your videos, I know the lower viewcount ones might feel less impactful but for me they've been really useful to hold up a mirror on how I can improve, and in these really tough times continue to try to make my way.

[@bonce]

It depends on the person, but my absolute favourite one is 'ok I'll lead with mine..' (and then I tell them my most favourite bug I've ever written) '....whats the funniest bug you've ever written or seen?'. Brings everything back to humour, reinforces the core of the creation process, allows them to slowly unfold into a more conversational tone and (if they want to) nerd out in this incredibly silly world we mess around in.

[@lpsowns]

Technical knowledge can be learned. Critical thinking, problem solving, communication, etc. are much more indicative of the quality of a candidate.

[@JGabrielRonkainen]

Could you talk about your experiences with hiring people from outside of the game development industry? What sort of challenges are there in getting an industry-outsider "up to speed" so to say when you hire them, and on the other hand, are there significant benefits?

[@seanothepop4638]

Also it would be good to see WHAT those jobs are. What jobs are in game design that are not technical? I imagine there's lots but would like to know.

[@veranish]

This would be super useful! I'm coming from making unreal and unity programs but for industry and research purposes, and having a really tough time getting a read on how interviewers and recruiters view that experience. It's seemed to run the gamut from "You are overqualified for this entry level role" to "You have absolutely no experience for this entry level role" haha, people vary but I'm looking for things I can do to more effectively communicate the overlapping parts.

[@FluffySylveonBoi]

My answer to the first question was (Spoiler ahead if you want to do it yourself, don't read this before you watch) I would make an option to insert the duplicate (or triplicate etc) at any position of the playlist the user wants personally, because some users love their songs in specific orders and the option would also contain the turn on/off button for duplicate songs for people who just want them once. You would also have an option to have all songs (both with duplicates and without) in orderly fashion, or just playing randomly.

[@theangel666100]

I would give the popup, but include a "don't ask again" button

[@ChrisHerborth]

Nice to hear you're taking a sane approach to this; I've been in a bunch (general tech industry, not games) where the interviewer's goal is to prove they're smarter than you, or if you don't know "the trick" to doing it right "right" way you're out. Your tests are more like behavioural tests, which are super important for figuring out how someone works on a team.

[@anonimowelwiatko9811]

Design questions were quite interesting, open and easy to talk through but as for programming, despite being SE for 7 years, I had no idea to check how it's implemented or even didn't think about meaning of these functions. If I want change int to string or string to int, I use existing functionalities. Then as for "if integer is even or odd" I thought first about modulo 2, into checking if last bit is "1" or "0" for number other than 0, into asking myself which bit is used to represent sign so I don't do do wrong comparison. For average, simple answer would be (a + b)/2. Then there is rounding or float, representation if we go for it, also if we add two ints, we have to make sure that we don't contain result in same format as we can get overflow. Returning reverse string by using reverse iterator was a no brainer but atoi/itoa really got me thinking. I didn't even think before that they meant ASCII to Int and reverse as I just had to use them, not engineer them myself. I had both division by powers of ten for each digit and bit operations ideas which probably could work. Really fun exercise, thanks for opportunity Tim!

[@gilgamecha]

I would agree that "role inflation" is the most common lie in interviews. You can drill into it by saying "what did you specifically do" but even then you sometimes get bold faced lies. Tests aren't a panacea but they are a useful cross check on forensic questioning.

[@LadislausMarguspa]

you can thank employers for training us to speak this way. if we dont exaggerate at all, we're not getting a call back to clarify hey what exactly was your responsibilities for X role from Y year... when we are honest and up front they just put our applications in the shredder.

[@Vanity0666]

@@LadislausMarguspa yep! The authority creates the rules to navigating the system which the authority operates. Just like how bad parents teach their kids to lie and hide mistakes in fear of retributive punishment so too do potential employees learn to lie to the authority in hiring in order to secure position and favor.

[@Postal0311]

There was a guy who worked in my team at one company. Later he applied for a position where he would work for me again at a different company. On his resume, he had put that he was the manager of the department we worked in at the first company. Funny enough, he still got the job. I later saw an updated resume of his where he put that he was the manager at the second company, but he had spelled the companies name wrong. He is a very hard working employee, I don't understand why he would put in false roles on his resume.

[@thelightside77]

Hey Tim. I'm here to say that i played The Outer Worlds yesterday with my friend. The same dude i gamed with for years basically starting with Fallout. The game is amazing. I love the humor in it, the dialogue is top notch and you can tell even by first visiting Edgewater that it can be very creepy and mysterious as well. It has been a while since i was drawn to a world like this. Thank you for these gaming experiences man. Without them we wouldn't be the people we are today🙏

[@mussdem]

The largest issue with interview questions like that ones you mentioned is that they in no way actually filter out people who would be bad at the job, they just filter out people who are bad at interview questions. I cannot do those kinds of interviews because I always get stressed as my chance of employment and income are on the line and I can't afford to make a mistake. This is in no way indicative of how I would handle stress on the job either as on a job I've always been able to leave the computer for 15/20 minutes and come back refreshed and less stressed ready to continue, that cannot happen in interviews. The amount of times I've seen people who interview perfectly come on to the job and are nothing but dead weight is astronomical, people study to be good at interviews not be good at the job. A much better way to interview is to ask them for a project they have worked on and review that code, or at least give them the option to do so, every job I've ever had has had that mentality.

[@liaminwales]

In academic circles there's reading lists you keep, there to show your well read and what subject you read etc. It's super common that people fake them, they read the synopsis and a review to pass off as there opinion on the book if questioned. It's a pain but sometimes you have to push to find if some one relay has read the book or understand the subject, it's that much of a problem.

[@user-yk1cw8im4h]

having to read in the first place is flawed.

[@aNerdNamedJames]

​@@user-yk1cw8im4h I don't know. I think it's be a benefit to game dev as a whole if, for instance, every game dev reads the game design book that coined the term "game feel"

[@Nikelaos_Khristianos]

I have heard horror stories about students who lie about what books they’ve read during their Oxford or Cambridge interviews… it’s such a risky thing to do! Especially because the person interviewing you has most likely written a paper or an entire other book about the book that you’re claiming to have read. That can become such an embarrassing situation.

[@aNerdNamedJames]

Odd as it may sound, this was probably one of the most reassuring videos I've seen in months, so, thank you very much for that!

[@RobLang]

I've interviewed more than a hundred people in my career (web, not game dev) and my approach is like Tim's. I'm trying to ask two questions: 1. What drives them? 2. If I locked this person in a room with a computer, how long would it be before they try and create something with it? The most valuable part of an interview is asking someone to demo and talk through code they're proud of. Loads of people can code but I need people who can review, criticise, explain, see improvements and question code. Even good juniors with the nastiest python can interview well. I'm trying to find out what they do know; not what they don't. I hope to push the passion button too!

[@CainOnGames]

Rob, I just read your Shared Pool System, and I think it’s a wonderfully creative idea to get players to actively have more fun and be more involved in the game. Now I’m wondering how to automate it for a CRPG….

[@ZTriggerGaming]

My favorite question I’ve been asked in an interview is, “Have you ever violated policy in the workplace and why?” This question alone landed me the job. First off, it challenges their honesty and experience. If you say you’ve never violated a policy, either you’re lying or you haven’t been in a high pressure situation that required a creative solution. If you say yes, you’re a person who will give honest and possibly critical feedback. Good leaders love this in an employee. Your why is still very important here. More important than following rules is your ability to articulate your reasoning for what you did. This method is a great way to find out if someone has this skill. That’s not to say an answer of “no” disqualifies you from employment though. While it does sound like a lie, the reason you give for your answer is pivotal, as it challenges your ability to articulate a reason even more. Great stuff. I really enjoyed your questions. Lots of valuable insight here and very smart!

[@BorrisChan]

So incredibly helpful. Working on the answers to your non-programming questions now as an exercise!

[@eggy5749]

have a great fondness of tim cain's videos. struggling with low self esteem and nervousness, watching these videos can greatly boost my confidence in what im capable of doing and learning to to work well with other and for others :)

[@thisisfyne]

Fully agreed here! At first when I was putting up a 3D surfacing team at work, we didn't do tests before contracts. Turns out that half the time, the artists didn't even grasp PBR principles despite having good portfolios. Even one guy with a GREAT portfolio couldn't follow common good practices and feedback. Needless to say this was a huge struggle for everybody involved, so we implemented a quick paid test (2-3 days of work in a context as if they were a member of the team) to confirm the candidates' skills and profile. I think if done properly, tests can be a GOOD thing. Even when that studio had hired me years prior they did so with a test, and it helped me better understand what they expected. The only scenario in which I'm against is if a) it's free labor and b) the company is still allowed to use it if they want. But I haven't seen that often, and when I did, refused.

[@SyndicateOperative]

2-3 days as a test is a great idea - it sounds like your issues were more about compatibility than skill level, though. 'Common good practices' is inevitably code for 'whatever we arbitrarily decide to do as a group', especially if you're getting multiple people that don't conform to them.

[@thisisfyne]

@@SyndicateOperative No no in some cases there were definitely skill issues, not just "incompatibility". There are universally true good practice when talking, in my instance, about surfacing, lighting, and rendering. I mentioned PBR (physically based rendering), which is a set of rules applicable to all modern 3D films and games that one *should* know to do textures and shading. Also for lighting there are long-standing key principles to know like visual shaping and mood/colour theories and whatnot. Those and many more were lacking in some candidates that we had tried, to my surprise. So really not just a matter of taste or anything; we're talking about actual competency here. Imagine trying to hire a car mechanic who wasn't able to change oil or tires. That's why we started testing.

[@GameMiestro]

I like the way you conduct these sorts of in person interview questions because they're really testing how people think and people's ability to communicate ideas and walk through problems. They aren't trying to trick people with "gotcha" questions or test people's ability to memorize obscure programming "tricks" and super efficient algorithms like you'd see in some Google programming challenge that would generally just be better off being researched online anyways. I think that's what some people are afraid of with these interviews, that they're going to be treated like a contestant on Jeopardy and they're going to have to pull all this programming trivia out of a hat when in reality being able to construct easily understood and well reasoned solutions to general problems in a language of choice is usually enough to get started in software development.

[@arcan762]

I suppose so many of us have been conditioned to lie a bit, at least when submitting an application. If the list of expectations is unreasonably high or it seems like a company is just looking for a golden goose of an employee, then guess what, people are going to claim to be that golden goose (regardless of whether they actually are or not). If being totally honest means that your application likely just goes straight into the trash, then why bother? I get it wastes a lot of peoples' time, but if the success rate for lying is non-zero (i.e. you can manage to bluff your way through an interview, or they just ask questions you happen to know good answers to regardless of what the job description specifically asks for), then why not just shoot your shot and see what happens? It is like clickbait, it works because it makes sense in a way. Why would you click on something that you otherwise wouldn't want to click on if it weren't clickbait?

[@ryanfriscia1132]

I had one the worst interview of my life because I could not answer a in-person interview question. It SAVED me from making a horrible career move and getting myself into the wrong field of work. It was a great wake up call that I could have made a major career mistake.

[@TREEKO978]

getting a notification to this video while OMW to a job interview (went well)

[@robshelby]

I've been coding professionally since 94. Im usually the lead. I still can't code on a whiteboard. My mind goes blank. At least give them an ide if you are asking for real code and not just thought patterns and diagrams.

[@joeruder]

LOL...for sure! I went to a interview a few years back and it was all white board and they were dinging me on syntax/spelling. I was like, I never get past the first couple letters before auto-complete takes it. Back when I first started (early 90's) I am sure I could write everything out as that is what I had to do all the time.

[@OpenGL4ever]

And here comes the big question: Would you give them a calculator if you give them a very simple math problem?

[@CainOnGames]

@OpenGL4ever Sure. But why would I give them a simple math problem?

[@OpenGL4ever]

@@CainOnGames Oh, i was asking robshelby. He asked for an IDE, so I wanted to know if he would also provide a calculator in a math question.

[@Xeit]

This interveiw questions (programming) were funny, had to think about them, thanks! During question 2 I was like "Yeah that's easy." Then I heard your question "are you sure you counted ALL possible numbers" and I immediately went "OhNo.png".

[@Tailmonsterfriend]

As an Editor working in the gaming industry, I've developed what we call internally an Editorial Interview Test, which is a take-home test we ask candidates who made it through the first round to take and return prior to their second round. You're given a simplified version of our Style Guide and a number of copy examples to edit. Like, "Here's the style guide, here's some copy, please edit the copy." The trick is that every piece of copy contains at least three types of error (often more) that can be any of the following: grammatical errors, typos, weak word choices, style guide inconsistency, echoes, inconsistent Capitalization, vagueness, etc. Then during the second round, we talk through each example with the candidate and get their thoughts. Sure you can have someone else do the writing portion for you, but the writing is actually the least important part of the test. What we really look for is how the candidate explains their writing and what their thought process looks like. No one has thus far aced this test (including other writers across the company who helped testing the test), but it's not important to get everything right. It's more important to understand how a candidate thinks. Interviewing is an art, and I like how Tim uses measurable (and rankable, so you can compare candidates against each other) questions to tease out hard-to-measure qualities like critical thinking, analysis, design literacy, etc.

[@joshbehrens9967]

From a usability perspective I think a good answer for the 'playlist editor' question would be to remove the old position of that song and re-add it to the top of the stack. The user feels the procedure did something, and the duplicate is pruned (or rather the song is moved to the most recent date added). I feel a popup is fine but intrusive, particularly if the user *forgot* that song was on there in the first place. At that point does the user even care about it's existence? Or, do they just want that song in an easy to find place? Especially considering typical listening patterns, where we listen to a small group of songs and things are added and removed mentally all the time. (unless of course the playlists order matters - in which case there should be a bool with a label that describes it is meant to be listened to in order. I'm sure there is some nomenclature for that)

[@iiropeltonen]

For at home tests, you could define a test and then face-to-face ask them to implement changes or a small new feature. Which they can do if it's their code

[@CausticTitan]

I think that take-home tests with a candidate-led review of the questions and answers is probably the best mix. Make the questions open-ended enough to allow for individual interpretation (just like how a work ticket might be open ended) and then ask them about what thoughts/issues they had while working on it, how they came to the answer, and what else they considered. This allows and maybe even rewards them for looking something up or asking for help (the same way they would while actually on the job), but if they do, they will actually learn it for real then and there. With this approach you get to see how resourceful they are, how well they learn, how well they take negative or positive feedback (if you give any)

[@steinmov]

I am an animator. However, this topic is of interest to artists as well as programmers and designers. An observation, people who exaggerate on their resumes get to the interviews. Being straight forward and honest about what I have done and can do does not get me to the interview. This is an observation from more than a few years working. I don't inflate my ability. Also, I have read programmers comments in various posts who work in IT positions. And I have friends who are programmers in games. They all say that looking up possible solutions is normal and in interviews they say that if they answer that they would not look up solutions there would be raised eye brows because a programmer just does not have all of the answers.

[@OpenGL4ever]

7:53 About b) and c). Should the words in the string have different lengths? It's easy with "how are you" -> "you are how". But the level of difficulty increases when the words are of different lengths. Example: "you are super" -> "super are you". Especially with the requirement not to use extra memory.

[@CainOnGames]

Yes. You can’t assume the words all have the same length.

[@OpenGL4ever]

@@CainOnGames Thank you for your answer.

[@oliorogue]

All of these questions are super fun. I wish any of my jobs cared about those things.

[@Goober_Guy]

I'm a amateur 3d artist and love this video. The level design question near the end i loved. I felt if i were asked them I would love to answer them in so many different ways as i do play games with analytical mind and am always asking myself what could be done better and how to balance areas of games with new items designs. Love watching your video and it makes me more excited to get into the game industry every time.

[@OpenGL4ever]

4:45 Okay, here is my answer to the music playlist: From experience, I can say that it has often happened to me that I want to listen to the same song several times. And I don't want to be asked every time if I really want to add the same song to my list again. So I would choose option B) 3:53 . Which means, the duplicate song is quietly added as the user wanted. What I would consider, however, is offering a button that can be used to remove duplicates from an existing playlist. The user would have to actively press this button once, however. Would that be a good answer for you?

[@CainOnGames]

I accept any answer. What I like to hear is the reasoning behind it, which you provided, along with a recommendation for an extra button.

[@OpenGL4ever]

@@CainOnGames Thank you for your answer.

[@ThePhiloctopus]

Heres my initial thinking on the playlist problem, going from 'least' assumptions to 'most/biggest' assumptions. 1. You want it on the list. this is always true, so to minimally satisfy this assumption we do nothing. 2. You want it on the list, and you want to see it at the end of the list (where you just 'put' it). We move the item to the end of the list. 3. You want it on the list, at the end, in addition to where it was previously. We duplicate it. My solution would start by implementing 1 and only move to 2 or 3 after these have been shows to have benefits that outweigh the cons, for most people. I.e. what are most peoples expectations. If we have to imlement 2 or 3, but also communicate/manage those expectations, then I would look at option fields, feedback, etc. Personally I prefer making minimal assumptions and avoiding complex behaviours - E.g. imagine allowing duplicates but not sequential duplicates. Do we have a separate option for sequential duplicates? LOL. Just keep it simple :D

[@alpha007org]

My first thought was asking how is the user adding the song. There's a difference, if he's clicking in file manager and if he's dragging from file manager to playlist or inside music player.

[@evadecaptcha]

Very interesting questions. As a programmer & a veteran gamer, they were fun to think about and answer.

[@EscelatesQuick]

I use one of three methods in interviews, sometimes all three and other times only one. It's worth noting that I typically do 90 minute interviews with the caveat that they'll usually finish early. I also don't really tend to spend time on multiple rounds etc. 1) peer review, present them with a sample of work that's meant to be complete ask them to review and provide feedback. The piece of work needs to be relevant to the craft/ role I'm hiring for - some examples of what this might look like are; a status report, a marketing plan, a complete but small section of a design document and small and self explanatory piece of code. There should be a technical flaw and a "philosophical flaw" baked in. I don't take it poorly if people don't notice the flaws but they should be able to have a material conversation about what they've noticed/ improvements/ following questions etc. Typically I'll give them 5 minutes alone time with the material to digest, make notes. 2) brainstorm, present them with a problem statement - similar to your playlist but open ended rather than optional. The scenario should be in a tangential industry and focus on collaboration over technical, I'm usually looking to see how they'll work with me and what level of certainty they need to operate. I've let this take up to 30 minutes alone for certain roles where autonomy needs to be high. 3) ask them for feedback on what they'd do to improve the interview process. This serves two functions, the first is it's good to have feedback - the second is it gives me a sense of how freely they'll talk to me about issues. If they can't offer candid and respectful feedback prior to signing a contract it's unlikely that will improve without coaching after signing. None of these in isolation have ever dictated an interview outcome, but 2 or more in concert tends to paint a picture. Regarding 1 & 2, the first time putting the resources/ scenarios for a given role might take some effort but it pays off over time. The biggest flaw I've noticed specifically for the peer review is some interviewers are more brutal than others so it's higher effort to use at scale without disqualifying good talent accidentally.

[@jakeoreilly]

wow it was fun actually thinking about each of those design questions! didn't know i had so many opinions off the top of my head!

[@Anonymous-ks8el]

When you ask "what is a feature you don't like in your top 3 games" is it acceptable to answer with how a feature was tuned (too little vs. too much) or how it effected the flow from earlier games And when you ask "how would you make a good combat environment?" how would you receive a response based on a genre & game you're not familiar with I could go on about how TTK & suppression in Battlefield 4 was unfairly tuned between classes (it was a tad too strong in BF3 but fairly tuned), or how a 'good' FPS map needs to be designed around how fast you want the gameplay to flow (pump & rest or solid pace the whole time) while using terrain, structures+props, destruction and elevation to limit LoS, possible engagement distance and how teams can manoeuvre on objectives, but you the interviewer would have no way of knowing if the response or my manner is a 'good' because you don't play FPS or Battlefield

[@Hjorth87]

I'm a gamer and tabletop player, but I know very little about game dev or programming (picked up a little in university, but I work in technical surveying), but your questions about game designs, classes vs classless and rating weapons ect made me go, hmm, I can actually answer that, and I would love to work with that. Especially getting into the nitty gritty of mechanic design

[@Khanemis]

These are some really interesting and fun questions. I would love to discuss such design choices with someone even without it being a job interview. So many choices, so many ways to approach the topic and delve deep into various pros and cons to each solution. Can't see why someone would dislike such job interview.

[@Filkolev]

Completely agree with Tim. In my company we have two interviews for developers. One is technical and we want to see what the person knows and what they did exactly, with details (without revealing anything sensitive of course). We want to see if they actually knew the code they were working on, how they contributed, how aware were they of their place within the larger organization, etc. We've seen plenty of inflated CVs, someone saying they did something, but when asked to explain they deflect with generalizations. The other interview is more behavioral. It is a live coding task on an almost complete small project. We want to see how they approach the code. Do they spend most of their time looking at meaningless details instead of the big picture? How they communicate with other people? Do they ask questions? What questions? How do they discuss technical decisions with their "junior buddy"? How do they respond to criticism of the code they have written? The question we want to answer here is: (regardless of technical skills) do we want to work with this person?

[@someguy999]

This was a really fun video and I was thinking of answers myself. For the playlist question, I was hoping he'd have an option for just adding the song, but to leave a little message on the bottom of the screen with an undo button. That way the user would be aware that they added a duplicate, but would be less inconvenienced if they wanted to add a bunch of them.

[@mememachine-386]

I generally don't mind interview tests (except for at home ones for all the reasons Tim mentioned). I can't really think of a better way to see if a candidate knows their stuff. However, I do have a problem when companies have multiple rounds of technical interviews or interviews that are more than 3 hours. If you can't figure out if a candidate knows their stuff and is a good fit for the team within a 3 hour session, you're probably not asking the right questions.

[@MontaguStudios]

I think I can handle all the questions Tim mentioned. The worst one is, "Why should we hire you?" That's an Uzi-wielding Ian kinda question.

[@txdmsk]

Once upon a time I interviewed with a leading tech company. I negotiated 100% remote work. F/k commuting. I want to spend time with my kids, not in traffic. They said that the work would be 100% remote, forever, guaranteed, for real. BUT, in the contract they can only give me 80% in writing, as "that is the only contract they have". I told them no thanks, but they convinced me that I can trust them. Well, 2 years later they decided to force everyone into the office for the maximum number of days allowed by their contract, so I quit for a 50% pay bump.

[@Mugnum_]

Whiteboard interviews can be improved with providing access to IDE of sorts. Some problems require fiddling with the code in an especially unusual manner (like converting said reversed string to ascii, etc.), something that interviewee doesn't necessarily do everyday. I guess it just depends on goals of these tests: - do you want to weed out the people who are pretending to have any programming experience, - or do you want to verify that an interviewee is considering various limitations and pitfalls to obvious solutions. An interesting spin on these tests would be "Spot the mistakes in this code example". The design questions are super interesting, definitely give lots of room for thinking, and makes me wanna see a room of professional designers brain storm their ideas. Though maybe it's just the nature of briefly mentioning examples of questions in a video, but I always felt like I needed to hear more context to these questions. Because answers (and question's goals) vary greatly depending on genre and that sort of thing (...unless that's the whole point of them).

[@fitemilkhero]

I like take home tests, and what ive done is after someone submits it. If its good we bring them in for an on-site, ask them questions and then ask to add a new small feature. Helps us know if they are familiar with the code and how they work on the spot.

[@Shoto_UK]

Really enjoy these videos Tim. Come away with new knowledge but also things to consider.

[@txdmsk]

Those are all excellent questions for designers. They show you how the candidates think. Cool video.

[@Postal0311]

I once interviewed a guy for a CAD position and I asked them what CAD software they had previously used. They replied that all the software is pretty much the same so they can use any of it. I clarified that I would like to know a specific software they had experience with, and they would not (maybe could not) name any software. Had they named ANY of the many CAD programs they would have gotten the job, but because they didn't they got disqualified.

[@txdmsk]

Once upon a time I applied to a job in a company, and 3 department heads came to the interview because based on my CV all three of them were interested in hiring me. The interview went well, all 3 of them gave me an offer. I told one of the managers that I am interested in a certain technology. The manager enthusiastically told me that of course! his team works with that technology all the time! So I accepted the offer with his team. Well, it turned out that the manager lied during the interview, as there was 0 work with that tech in the whole team.

[@dustyyyyyyyy7]

It seems like all of those questions would make for a great conversation overall. Even if I didn't get the job, the interview will have been enjoyable at any rate.

[@spudd86]

I hate the what is the hardest x you've ever had to do type of question. And even tell us about a time you had to x. I always draw a blank. Even trying to prepare to a be able to answer them. My episodic memory just doesn't work to retrive memories with that kind of prompt. I need something more directly related, like a similar problem. I know this is a me issue. But I'm probably not the only person who's memory worns this way. I don't mind sketching things out on a whiteboard. I actually like the playlist question. Though given that nearly every application I've ever used does the same thing I'm.not sure about it's utility. They all just let you add the duplicate song, since it's a "play list" just a list of things to play. But as a question category I like the idea. Honestly for a basoc coding test puttong someone in a room with a provided laptop for two hours seems like it might work. Lock it down to only a few things, like docs and give them a data structures book. I frequently check myself on algorithm choice to see if there's a better one, google is much better for this than a book, since csn just throw some terms like dag positive weight mst and get the optimal algorithim. I'm.not sure how you could lock it down enough to block them potentially getting help and still giving google, so book.

[@txdmsk]

The return whether integer is even or odd question is easy! Just create an array of integers where you input all the odd integers and make the code go through all the array and return true if the picked integer is same as your input. Then do the same with an array of even integers!

[@OpenGL4ever]

Your solution is very slow and requires a lot of memory. Don't do that! Use the modulo operator. (i % 2). If the result is zero, the number was even, if the result is not zero, the number was odd. That's much faster and it requires only a few bytes of memory and can be done directly with CPU registers.

[@txdmsk]

@@OpenGL4ever It was a joke / autism check. Obviously it would be rather silly to process an infinite number of numbers.

[@txdmsk]

@@OpenGL4ever Can't tell if you are trolling or autistic, but whatever.

[@danielszemborski]

I'm surprised the creator of Fallout doesn't give his interviewees the G.O.A.T. exam.

[@ItalianRetroGuy]

Regarding the playlist: I would ask the user (popup) whether they want to add the duplicate or not when adding the song, with a checkbox to remember the choice on that playlist. Perhaps there could also be a little extra options panel when creating a new playlist where the choice can be made proactively and avoid the popup later.

[@thatsagoodone8283]

As someone who had many job interviews (as the interviewer) I really enjoyed interviews for coders the most. You can ask questions and give examples and cut all the BS that you normally get in job interviews. As an applicant, it is your task to sell yourself the best way possible to get most money for it. It is the task of the interviewer to be very skeptical and dont overpay for what you are getting. All in good fun and with manners, of course. But someone heavily specialized in taxes ... this is not that likely to pull of. Same with lawyers or other jobs, where research is actually a valid approach to the problem. You can come up with a riddle to see the way they are thinking, but truth be told, coders are better in articulating an interative / explorative approach to solve a problem. Business Administration often goes into very vague answers that could be anything. Kinda happy that I dont have those interviews anymore. The best were btw youngsters (ages between 16 - 19 yo). They often lack a filter and were not as good in giving the estimated answers. Very refreshing and both sides got what they looked for. Because if you know, that they are inexperienced, you can lead them way better and be direct/honest in what will be offered.

[@MrJoking4fun]

I don’t have much to do at work today. So I have spent no kidding the last 5 hours doing very little actual work and instead spending my time imagining how I’d respond to them in my head. I love video games :)

[@The-cyber-imbiber]

Instructor: "Give me the average of two ints..." High level languages, "Easy. There are no ints, doubles, floats, etc. There are only numbers." Low level laguages, "The game crashed because the sum of the two ints were greater than the maximum size of an intiger. I converted them to doubles but it still doesn't work because I can't do division with doubles. Do I need to use a the equivalent to "double" but for floats? A floutble? Maybe I should have just stuck with python."

[@vast634]

in C you can use long for the intermediary sum: long sum = (long)num1 + (long)num2; return (float)sum / 2.0f;

[@The-cyber-imbiber]

great take@@cat-le1hf

[@perthhi1]

I agree with you in general, except in this case the question IS the details. The hint is that literally everyone knows how to average 2 numbers. The solution is to avoid adding the numbers initially: ((a - b) / 2) + b@@cat-le1hf

[@bratttn]

When I’m interviewing a candidate after the person presents his background, I usually start by asking to describe a single business process, for example, I press a button, what happens, then I stop him at certain technological nodes and go deeper, ask for alternative solutions, specific improvements etc.. Not every person makes it through the whole round trip.

[@SiriusMined]

Exactly. Some people are good at interviewing, and not good at the work as they say they are.

[@RaccoonKCD]

6:25 As somebody who has absolutely zero programming experience these questions gave me anxiety for a field ill probably never enter

[@ArthurKannibal]

This video made me realize that I would be focusing on impressing tim cain as I answer the questions. Never wanted to ace an imaginary interview this badly.

[@azx7023]

There's two big interview questions I would use you don't mention here, and that's fizz buzz, good for making sure programmers can work with abstract tasks, and asking them to solve a specific problem in a specific way. This both tells you if they're a good problem, solver, and if they take a holistic approach or just charge for the first "solution" they see

[@qpid8110]

My first instinct when seeing the title of the video. "You're in a desert, walking along when you look down and see a tortoise. It's crawling toward you. You reach down and flip it over on its back, its belly baking in the hot sun, beating its legs trying to turn itself over. But it can't. Not without your help. But you're not helping. Why is that?" 😂 I'm so looking forward to the video!

[@SPTX.]

I go for solution E. Add the duplicate, highlight the song(s) in the list with a color, such as yellow to warn about it (also make it verbose somewhere with same color coding). If the playlist isn't shuffled sometimes you may just want to have the same song repeat at different times to set a mood. I hate pop ups. They waste everyone's time.

[@pieflies]

I like your questions. I’d be happy to be on the receiving end of an interview like that.

[@StefanHayden]

great video! I know many people hate take home exercises but some do like them. Many also hate live coding on white board. It's almost impossible to make everybody happy. I have conducted interview pipelines that gave people an option to which they would prefer. There are challenges with take home exercises but I do think they can be overcome. To make a take home work I think the trick is that you treat it as if they just wrote it on a white board and you have them walk you through it and explain everything, what choices they made, what they tried, and what didn't work. Maybe they had to look something up and learn how to do something new and that is also a valuable skill to demonstrate. If they got someone else to do it they should faulter in explaining their decision process. Would love to hear if anyone come up with new interview ideas you thought were interesting.

[@abrahamdrinkin2534]

When my father would interview people, he would always ask a few questions he knew the answers to just to see if someone would lie. And he would even ask follow up questions to see just how big they spin the lie.

[@scottishrob13]

I guess my problem with simple interview problems like that are that I'm never going to ask someone to do those things. I think the probing style of questions you outlined for designers are perfect for programmers as well. What's important is how they approach solving a problem, and how they react to having their solution challenged. Something high level, applicable to the role they'll be filling. Bonus points if it's a problem they'd be working on when they start, or one that has already been solved. Maybe something that would take a week to implement fully, but I want to see if they can come up with good angles of attack. There's still room for general questions that dig into technical competency too, but I don't think they need to take the form of "solve this trivial thing under pressure in front of us in an unfamiliar environment with correct syntax" haha It should be pretty easy to ask a few questions and get a sense for whether or not they actually know the basics.

[@rusty7448]

Now I really want to do a mockup interviw for a gamedesigner just to see if I'd be accidentally hireable. :D

[@lonneansekishoku8288]

*Puts one of Tim's video in the background while doing something else.* "Here's a list of questions for you!" *Video becomes the primary objective. Must answer every question!* I swear the design questions got me fired up. I'd have to pay attention to not go on a tangent as I explain a mechanic. XD

[@ch3rt.]

As a user, I don't like all those options for music playlist editor. 🙂 I would silently add small 2 in a circle to the line of a duplicate song, with simple hover-mouse info that this is a duplicate song. Then two ui buttons, one to "delete all duplicates" and a button "accept all duplicates (don't show a number)". These two buttons doesn't have to be visible on first sight, they could be kinda "level 2" of buttons. Of course it could be even more adjustable, but this would be my bare minimum - don't block the things user wants to do, try to provide options, but don't overcomplicate it.

[@PompousDingo]

Really liked this video. All of the questions helped me gather my ideas and opinions and it helped me see where my shortfalls might be

[@Nlelith]

I would much prefer to have an in-person test like that than a remote one. How many candidates do you usually interview with this approach?

[@denisecastellanos4866]

Honestly, were I hiring for the development of my game, I would circumvent the nonsense of job interviews and just build sincere relationships with people. ACTIVELY building TRUST with people who can be potential job candidates. Possibly dipping into modding communities. In modding communities you not only get to see what people are capable of, but depending on how they handle comments/criticism. You get some idea of their character as well. How interviewing is conducted, and especially how candidates are dealt with after the fact, tend to be riddled with bias, a level of inhumanity, and coldness which I would not want to keep perpetuating. I definitely feel that the system needs to change. The culture of distrust in the games industry I think causes more harm than good, in the long term.

[@NikoNOJ]

yesssss. 4k quality upgrade. More pixels on this dashing man!

[@brandonross7749]

My first thought was just have the already saved song just moved to the spot the duplicate would be added because they forgot about it being there and it would be easier for them to find it or they didn't like it in the order it was added the first time.

[@squib308]

Microtransactions! Everyone's favorite

[@fredrik3880]

Playing Fallout New Vegas again. What a great game. Love how they use the ambient music from f1 and f2. Just did there stands the grass. Amazing quest and vault!

[@Vanity0666]

I started writing out the the bool int true false statement and it took like 10 seconds longer than usual, that's a fun problem. The easiest way to do it would be to check for divisible by 2 for all evens, and if not divisible by 2 then they're odd, because int doesn't include decimal values.

[@talon4107]

I came across an approach with home test, after which on a in person interview we were talking about my solution. Why I did that, what I think could be improved etc. You're going know to if I did it by my answers. I also think some simple question like you mentioned are required (like how would you do x). I don't get it why people hate them.

[@fasgamboa]

Those even/odd questions remind me of my C classes 😵

[@danfors1333]

at 4:10 solution D, I said to my self for fun "crasch it" and by freak accident my Firefox crashed. Seems I was given instant bad karma for that evil answer.

[@hasha555]

Thank you for giving such specific examples of questions for interviews. I have an interview in a few days and would love some example answers that you would consider good or bad as I have an answer to all of them, however I'm always unsure if Im missing out on something or im not concrete/detailed enough.

[@Tenzarin]

Remodel your interview room into some kind of escape room where you walk them in and just never come back and see how long it takes to to try to escape. Tests reactivity.

[@Tom-f4d6l]

Now when people are doing programing tests you also have to look out for if they are using chat-gpt for their answers (instead of just google). My default answer to most programing test questions is "first I would google it because that sounds like a solved problem and someone smarter than me has probably has a good answer". However in game development I have generally found that most tutorials and examples are not particularly made to integrate into a larger system, so google is less helpful. I also like the game design questions, but I do wonder if chat-gpt would pass your design interview/test.

[@EB-cz4te]

I would absolutely love to do an interview test like this in theory, (I actually wouldn't mind being considered either) to a game development job.

[@mxdev6387]

I do embedded C, and always throw the averaging two unsigned ints as a soft entry question. Turns out it's not soft at all and it's 50/50 someone gets it right away. I start getting worried when you can't even lead someone to the answer.

[@nikital.6523]

Now this is how you do a call to action, heh. As someone that was on the both side on the interview process, I loathe the home assignments. When you the one doing the interview, it somehow feels even more of a waste of a time than when you're the one being given such a test. I don't think that interviews are going to go anywhere anytime soon - not at least before we become capable of violently ripping personality dump out of interviewee's mandatory neurolink or somesuch. I feel that with the most of the second half of the questions I'd start answering with "well, it depends..." since systems in any given game rarely exist in a vacuum. I'd still naturally narrow it down to a concrete answer of some kind since *almost* any solution is better than no solution - but always with a remark that actual conditions would influence such a design - perhaps, even in a dramatic fashion.

[@MrHarumakiSensei]

People lie at job interviews for the same reason they lie on dates. If they are honest, they will be passed over in favour of a liar. Maybe six months later the boss/company/girl/boy realises their mistake, but they'll usually start the interview/dating process over with new candidates. Very rarely will they go back and look for the original person who answered honestly.