WEBVTT 1 00:00:24.640 --> 00:00:25.960 Hey, Ben. 2 00:00:25.960 --> 00:00:26.880 Hey Matt. 3 00:00:26.880 --> 00:00:41.490 So there's a lot of things going on in the world right now, uh, in the tech world particularly, um, some of the FAANG companies are downsizing and we are at least the little company we work for one of the few places that are actually hiring at the moment. And this is, 4 00:00:41.490 --> 00:00:41.760 That's right. Yeah. 5 00:00:41.760 --> 00:00:45.860 not officially supported by our company in any meaningful way, but it, we have definitely 6 00:00:45.860 --> 00:00:46.940 This is not an advertisement. 7 00:00:46.940 --> 00:01:21.600 This is definitely not an ad advert. Yeah. It's not an advert. But you and I have been doing a lot of, uh, well, I say you and I, you have been doing an awful lot of resume filtering and, uh, interviewing and then interviews. I seem to be at the end of the interview pipeline, which means that I get the, the sort of the, the ones that have already been filtered through the folks that have already passed a whole bunch of, uh, other interviews. So I get one view. You are doing a lot of earlier stage interviews, but, I figured we should talk about that a little bit. And I know you have opinions and thoughts seeing as, all you seem to do is read resumes these days. 8 00:01:21.600 --> 00:02:06.140 Yeah. Read resumes and interview people. It's like a good, it's like a good third of my job these these days, which is not where I want it, but it's sort of like, you know, we are in this very fortunate position of, you know, having a company that is looking to hire and, you know, the rest of the world is, is not like that. And sort of that, you know, creates some opportunity for us to maybe, cuz you know, my, my perception, I, you know, I've never worked, you've worked for Google, I've never worked for like a big tech company. Right. Um, with like, you know, tens of thousands of employees and stuff like this. Right. Um, but certainly my perception of these layoffs is that, you know, if you're gonna lay off 10,000 people and you're gonna make the decision to do that and execute it in like a couple of months, which it sort of seems.. 9 00:02:06.140 --> 00:02:08.240 Seems to be the timeline. Yeah. 10 00:02:08.240 --> 00:02:26.940 Like, you're not making really fine grain decisions about which people you're, you're letting go and which people you're not. You know, you're just gonna cut whole departments and you're gonna be like, there's gonna be some great people in here. And maybe we could have retained them if we wanted to do this exhaustive re-interview process or some other sort of thing. But, but we're not gonna do that, we're just going to let the whole department go... 11 00:02:26.940 --> 00:02:30.640 By large, they have to use broad strokes to, to, to, yeah. 12 00:02:30.640 --> 00:02:31.340 Right. Right. 13 00:02:31.340 --> 00:02:58.140 Which is why you get some of these terrible stories about folks who've been there a long time, or they were like, uh, in the labor ward at the time when they discovered that they've been cut. Which means that like, no one really thought about it. Cuz there's, I don't believe that even in the most capitalistic society, anyone is that horrible to humans. Yeah. It just is how it came out because there's, it, it's big swathes of companies and without the personal touch that you would otherwise hope to have, especially after a long time in the company, but. Nevertheless, that's where we are. 14 00:02:58.140 --> 00:03:36.860 Yes. And that is what's happening right now. Uh, and so, you know, this is a sort of brief moment in time for us where while all of these layoffs are happening, some of which might be kind of arbitrary, and therefore laying off like very talented and experienced people. At the same time, we are trying to grow and sort of, there's this like, brief window where it's like, oh, well we might have an opportunity here to get a whole bunch of great people. And so that's why I'm spending a third of my day, reading through resumes and doing interviews and other things because I fully expect that, you know, six months from now or whatever, when the market has kind of settled out a little bit, that it will sort of be back to normal and we'll be, you know, 15 00:03:36.860 --> 00:03:44.240 I mean, let's hope so for the wider economy. Let's hope so. Even though it's lovely to have the choice right now, for us. 16 00:03:44.240 --> 00:04:30.800 Yes. I certainly, I think would be wise to operate under the assumption that this is going to sort of return to the way it was maybe not six months ago or a year ago when it was, you know, particularly hot, but like, you know, the way it was like two or three years ago. Right. Like, when we would, we would still sort of have to look pretty hard to find. We didn't have this huge influx of candidates, um, you know, take advantage of the opportunity where you can. So it has definitely been very top of mind for me. I've been doing a lot more interviews. I've been setting up new interview pipelines and trying to figure out how to do that. Talking with recruiters about how to, you know, get the best candidates and how to filter them effectively while also trying to keep in mind that like, we have other jobs to do, . Uh, we have a bunch of stuff to build and, um, everyone involved in this process is also doing those other things. So everything we do is a, is a trade off here. 17 00:04:30.800 --> 00:05:14.920 Yeah. Yeah. It is. But yeah, it is, it is. I mean, there's even the old trade off of like, you know, hiring another person. I think you said it well when someone said, uh, to you, can we just hire some more people? And then it'll go faster, and you're like, though, I can show you a book that will tell you exactly why they're the opposite thing will happen. So, you know, there's that balance as well. Right. You know, the time that we take to interview folks is lost time in some way if they don't work out. Um, it's important time, but it's, it's not time that you are building stuff. And when you hire someone new, you have a tax on you for a while while they come up to speed and while you get them the opportunity to, to grow and become more of a net positive than somebody who is asking like, where's the bathroom? I don't even know where the, how do I go ? How does the coffee machine work? Ah! 18 00:05:14.920 --> 00:06:02.740 Right. Yes. Yes. So, well, and our coffee machine is particularly complicated. Um, the, uh, yeah, I mean, and, and I think one of the things that, that everyone looks at when they're, they're looking at these like a, a potential candidate, um, is, uh, sort of like how long it's gonna take you to get them up to sort of that net positive level. Right? And for somebody that's right outta school, like you're making more of an investment in them. And it's gonna take a longer period of time, but for somebody with more experience, like, you know, part of the reason you're hiring people with the experience is that you don't have to teach them as much, and they should hopefully get to that point sooner. But it's true for pretty much every, I've never had someone that joined a team that I was on, and like day one it was like, oh, we're getting more out of you than we're putting in. Right. Like, that just, it just doesn't. I've never seen that. 19 00:06:02.740 --> 00:06:04.380 It's doesn't happen, you know. Yeah. 20 00:06:04.380 --> 00:06:50.100 Sometimes it's only like weeks, but it's never like the first day. So, um, you know, and, and then with, with people who are right outta school, like I expect that process to take months and sometimes even years to really truly get to the point where it's like, you are producing the level of quality that I would want and expect for this particular type of project. It can vary, um, and, and you're doing it, uh, you know, at a level that means that, you know, for every two hours of guidance and instruction and peer review and all of the other forms of input that I'm possibly giving you, I'm getting more than two hours out. Right. That can, that can easily take months to get to that point with somebody who's right outta school. 21 00:06:50.100 --> 00:07:29.440 Right. Understandably. I mean, like, that's it, it, there are whole processes that are vary from company to company. I mean, you and I are quite lucky that, you know, we've, we've found company, well, we've worked to at the same companies, and so therefore we've brought the practices that we know and love with us. So we kinda like, Hey, this is a familiar process. I wonder why Oh, cause we invented it at the last company, . Right, right, right. But like for somebody who's just fresh out of school, um, just even like doing pull requests and understanding how that process works or, you know, how do you design things in a way that is, uh, or what are the things that are important in a piece of software that you're not just gonna hand in at the end of a semester and never have to look at again, but it is gonna haunt you for the next six years. , 22 00:07:29.440 --> 00:08:11.220 You know, what do you, how does that change the way you write software and all of the things in between there? So, yeah, I, so what kind of, I mean, how does one go about for say, let's say a mid to senior person, not the, the new grads that we just talked about? Cause I think we can all see the shape of what that looks like, and we all understand some of the things that they're gonna, um, perhaps not have been exposed to, but um, you can get somebody who's been working for 3, 4, 5 years. Um, and then how do we look for the things that are gonna say it'll be weeks rather than months before you are, uh, we're getting more energy out of you than we're putting in, you know, some fusion reactor of your, your brain will be net positive 23 00:08:11.220 --> 00:08:25.360 Exactly. Exactly. I mean, so I think that we do it a little differently than a lot of other firms do. It. I get a lot of feedback from people that had been through our interview process and they are like, wow, this was really different. Like, you guys didn't do the regular sort of like, leet code, 24 00:08:25.360 --> 00:08:26.160 oh gosh, 25 00:08:26.160 --> 00:08:54.140 hacker rank whatever, silly. You know, I know those things are so gross, but it's like, you guys didn't do that. You did something very different. Um, and it's like, yeah. Like, like, you know, what we do, um, I, I think is actually kind of unique in that like our sort of, you know, our, our process is pretty normal at the beginning. It's very typical at the beginning, you know, someone sends us a resume. We have somebody call 'em up and we say like, Hey, like, where, where do you live and where do you wanna work and what have you done? And just sort of, you know, A, make sure that they're a real person because you know, we've, had that problem sometimes. 26 00:08:54.140 --> 00:08:57.880 We've had that! we found. Yeah. 27 00:08:57.880 --> 00:09:50.660 Uh, and then just sort of generally get a sense from them of like what they're looking for and do an introductory call. And then very soon after that, usually immediately after that, we send them a programming test. And this programming test is open source. It's on our public GitHub. We share it with everybody. And part of the reason that we do that, and, and I know, I don't know if you've had the same experience, is, you know, finance can be a little bit of a, of a, of a, of an old boys club, right. Where it's sort of like, people know people and you get these things of like, oh yeah, the interview process over at Jump is like this and they're gonna ask you these kinds of questions. And that's like, I think profoundly unfair. It's sort of like you're just giving an advantage to the people that are already in the industry. Beyond their industry experience. Like, this isn't, this isn't industry experience. It's not like you have depth, you know, in-depth knowledge of how equity markets work. And that's what's giving you... 28 00:09:50.660 --> 00:10:07.240 You know Barry at whatever. And they say, you know, they say, well, well this is what they asked me when I came in on the way in. And, you know, and there, I mean, there are websites nowadays that share some of this stuff and information, which is kind of another thing where it's like, well, if that's gonna happen anyway, let's just put it out there in the open. And then it's, then we've leveled the playing field for everyone. 29 00:10:07.240 --> 00:10:20.060 Exactly. Exactly. We're leveling the playing field for the people who, uh, might think that like, you know, digging into those things might be slightly, uh, not dishonest, but just like circumventing the process. Right. 30 00:10:20.060 --> 00:10:20.220 Right, 31 00:10:20.220 --> 00:10:22.020 Right. And it's like, no, no, no, we're gonna make sure that this is 32 00:10:22.020 --> 00:10:42.940 I could imagine myself feeling that like, well, I've noticed, oh gosh, on even like this. Right? Oh, I just noticed that they've, maybe they've didn't know that their, their, their, uh, interview test is, is open source and publicly available. Uh, I won't look at it. Right. And then that would be like my thoughts about the process. But as you know, but no, we're doing it on a purpose. It's there so that everyone can see it. Ahead of time. 33 00:10:42.940 --> 00:11:40.900 Exactly. So we have this, this very open process, uh, and the, you know, people ask people to do the problem, they do the problem, they submit it. And then the follow on to that is also, I think, pretty unique in that like a lot of what the, the follow on interviews do are their focus on programming and the programming process. So something that I, that I see a lot with kind of like technical interviews and programming interviews, especially like pair programming style interviews, sort of dysfunctional version of pair programming style interviews, is that you're basically just watching somebody code as quickly as they possibly can. Right. It's like, how much functionality can you cram into a 60 minute, or 90 minute or two hour, god forbid, programming session? And they've got this like, you know, checklist of like, were they able to implement this? Were they able to implement this? Were they able to implement this? And if you get all the way through the checklist, then you pass the interview. And if you don't, then you're not fast enough and you fail. And it's like, well, if that's what this job is gonna be like, well then just fail me now because I don't, 34 00:11:40.900 --> 00:11:42.200 I don't want that job. 35 00:11:42.200 --> 00:11:48.120 Right. I don't want to be writing code as quickly as I possibly can for two hours straight with somebody looking over my shoulder. 36 00:11:48.120 --> 00:11:52.560 I was gonna say breathing down neck the whole time. That's not a normal way. I don't write it that way. 37 00:11:52.560 --> 00:13:01.360 Yeah. Right. So, so we do the sort of like, you know, pairing style, collaborative style programming interviews, but the purpose of that is to evaluate their process, not their result. Right. We're not looking at like how much code they can crank out. They're looking at like, how do you think about software design? How do you think about testing? How do you think about performance? How do you think about data structures? Can you communicate these ideas to other people? Can you take your ideas and put them into code? Right. Can you manipulate code in order to make your ideas real? Right. And it's not about like, oh yeah, you implemented the three sorting algorithms, but not the fourth one, so you didn't make it. Right. So I think that's pretty unique. And I definitely think that gives us a good sense of like, in those early days when someone has just joined a company and they're gonna like come and they're gonna work with me and I'm gonna be like, let's work on this problem together. I'm gonna probably know in the first day what that trajectory is going to be over the next few months. I don't know if you've sort of had that experience of sort of like, yeah, after the first day it's like, yeah, this guy might pick up the a little slower, or, or, you know, that's, uh, 38 00:13:01.360 --> 00:14:04.630 It's definitely something, I mean, this is more an interview thing as I've often you've been, they've been, by the time I see a candidate, they've been through the, the various filters that, that come naturally out of the, the programming, uh, side of the test, uh, of the interview I should say. Uh, but yeah, definitely something I like to do is find something which is kind of on the border of something that somebody might know, um. But like, it's definitely not critical. So, um, I dunno if they say they have Java experience, they say, oh, tell me a little bit about what it means to, you know, Java versus say, uh, a real compiled language. And I know like, you could write a book on the answer to that question. Right. That's the thing. Right. Right. But I wanna push 'em until I get, they get to like an obvious point where they don't know anymore. And then if I know something, I'll maybe say, oh, this is how that works. Now, often it's, you know, some minutia of the garbage collector, or it's something in Python about how reference counting works or anything like that. And I'll try and teach them something just two minutes. Two minutes, you know, nothing. I mean, obviously I like to do that anyway, that's fun for me. 39 00:14:04.630 --> 00:14:05.140 Right, 40 00:14:05.140 --> 00:14:34.940 Right. Right. But then I try and make it so that later in the interview I call back and I see whether or not they can extrapolate from the new information that they've been given. Oh, does it work the same way as the thing you just taught me over here? And I'm like, oh, it does actually. Yeah. Great. You know, and that's, that's a very good sign for me that they can pick up things on the fly. And now obviously it's difficult cause interviews are under pressure. So it's not a necessarily an out that they don't, but it's a good sign for me and it can contributes to a positive signal from a candidate amongst all of the other things. Right. But 41 00:14:34.940 --> 00:14:36.160 Yeah, 42 00:14:36.160 --> 00:14:38.640 It's difficult, right? 43 00:14:38.640 --> 00:14:58.420 Mm-hmm. mm-hmm. , and I mean, inter interviews are just inherently a kind of process where you're just gonna get a lot of things wrong. Right. You hopefully err on the side of not hiring people who aren't a good fit and, and therefore reject people who might be. Like, that's the sort of smart way to do this. But it kind of sucks for everybody. 44 00:14:58.420 --> 00:15:34.840 It's, it's, It sucks if you're the candidate, right. It sucks. If you are the candidate, you are like, well, I was on the fence and, and know, and I think, you know, we've all seen candidates that have either we've, uh, we've passed on and then have thrived. Outside. Or yeah. You know, you've, one has stuck one's neck out for somebody because you are the one person who gave them the, like, the five star thumbs up. We have to hire them immediately. And everyone else is like, man, not really sure. And, you know, you sit your neck out and then they, they get hired and then it's like, oh my gosh, how do we, how, it's amazing how we got on with before this person was hired, right? Yeah. And that, that's, that's a nice feeling. But then you, but there's always like one quadrant of the, the 45 00:15:34.840 --> 00:15:35.700 Right, right. 46 00:15:35.700 --> 00:16:13.040 Uh, the thing that you don't get information on. So you don't know about the unknown unknowns. What, who, what, what about all those people that didn't interview very well at all, and no one stuck their neck out, and then you didn't hear about them ever again, but would've been amazing. All right. And it's really, really hard to get that right. But, you know, ultimately you've gotta protect the, your, the business at one level and also the working environment that you're gonna be bringing a new person into. And so you know, although it's unfortunate for the candidates, um, maybe if you're a a maybe it, it, it, you, you, you know, you have to round down and say, well, I'm sorry, this does not gonna work out this time. I, I, yeah. But it's never, yeah. I hate it. Everything about it's awful 47 00:16:13.040 --> 00:16:34.380 I, everything about it is awful. I totally agree. And, and it does kind of work out sometimes to be this sort of like subtractive process where you're putting people through these various examinations and all along the way you're expecting sort of like, yep, yep, this is good. Yep, yep, yep. And you sort of get to the end and it's sort of like, well, we've ran out of tests, so we're gonna hire you . Yeah. 48 00:16:34.380 --> 00:17:27.640 And that's, you know, nobody really wants to be like, you didn't do anything wrong, therefore we hired you, but But in a way, I mean, that's the, again, I'm at the end, I'm typically towards the end of the interview pipeline, just because of the way things are set up. So I have a very artificial view of these incredibly great candidates that effectively. That's how I feel. I feel like, like they've already come to, to me, I read through all the notes of everyone, and I'm like, everyone loves this person. I, I'm basically looking to see whether or not I can work with them, which is an important thing. Um, and I'm all the things that we just talked about in terms of their learning or whatever, and then I'm just here to say, yeah, everyone else was right. Great. Let's, let's bring them on board or whatever. Um, and I don't get a great way of calibrating how good that person is compared to all the other good people. You know, like if, if you've sifted out like 95% of the people, and I'm always seeing the top 5%, then I'm going, am I just too lenient here? Are, these people all great? 49 00:17:27.640 --> 00:18:09.820 Well, I think a great, a great antidote to that phenomena that we're talking about here is this sort of, yes, but not for me. Right. When you get feedback from people in an interview process, like, I think this person would be a great addition to the company, but I don't want to work with them. Right? Like, and that's kind of like rounds down to No, in a way. Now that's not universally true. There are definitely situations in which you have like very specific fits for people that would work out better in one area and another area. But at least in my mind, when I'm sort of looking at that, in that feedback from the group, after the interview process is complete, like there better be at least one person and hopefully multiple people that are like, I would be happy to work with this person. 50 00:18:09.820 --> 00:18:11.760 Or even I would champion them, you know, if 51 00:18:11.760 --> 00:18:12.560 Exactly. Like, I will advocate for them. 52 00:18:12.560 --> 00:18:16.580 Yeah, I would override a No somewhere along the line for this person. Yeah, that's a great signal. 53 00:18:16.580 --> 00:18:49.180 Exactly. Yeah. Yeah. And if you don't have that, you might have fallen into this trap of, it's not that the person is actually a really great fit, it's just that all along the way they were like, well, I guess good enough, I guess. Good enough, I guess. Good enough. And at the end it's sort of like, well, I guess we need to give them an offer now. I don't know. Like, that's not a great outcome from that process. You need to have something I think, a little bit more, you need to have an advocate at the end of the process to say like, you know this, if no one else wants this person, I will gladly take them. Please, please, please. Can they be on my team? 54 00:18:49.180 --> 00:18:58.180 Right. And in fact, I see you a very, and on a recent candidate that went through our pipeline, you made that exact point, , which I thought, so you're putting your money where your mouth is quite literally here. 55 00:18:58.180 --> 00:19:12.000 Yeah. Yeah. That's true. That's true. Um, but yeah. Yeah. I mean, and I mean, beyond that, like, one of the things that I think there's a lot of economic value in is making interviews both humane and really fun. 56 00:19:12.000 --> 00:19:13.320 Right. Um, for certain. 57 00:19:13.320 --> 00:19:58.730 We, For two reasons. One is people, you know, interviews are just super stressful and, and especially engineers. Like we're hiring them to be able to think sometimes think very deeply about very complex problems. And an interview is not the sort of emotional state hopefully, that you're gonna be in when you're working at a job. And so, you know, like it's a, it's like a very well-known thing. You know, people under stress, their, their, their cognitive abilities drop. They, they stop being able to think of more creative ideas. I, myself, in my career have had multiple situations in which I've had some ongoing production issue, and I log into an, ssh into a box, and I just get dumber. 58 00:19:58.730 --> 00:19:59.640 Oh my gosh. 59 00:19:59.640 --> 00:20:10.640 Like, I forget the names of commands. I like start transposing things, and I have to double and triple and quadruple check the things that I would normally just be able to do without thinking. 60 00:20:10.640 --> 00:20:50.920 Right. Um, and that kind of stress absolutely happens in interviews, and it can really distort your picture of a candidate. And so I think one of the things that's really important as a way to, to sort of make, create a little bit of an antidote for that, I mean, you're never gonna be able to completely do it, is you have to think about your interview process and, and how you're gonna make it interesting and fun and engaging and something that like, is a challenge, but is also, you know, getting your mind into that mode of solving interesting problems and talking about interesting things as opposed to the like, oh my God, I'm, I'm, I'm making a fool of myself. I, I can't remember the answer to this question. I used to know this. Oh my God, what's happening? Right, like that's terrible. 61 00:20:50.920 --> 00:21:20.100 Right. You, I know you wanna set up situation where you're like, uh, this is something I go to pains to say, like, you know, we understand it's a stressful place, so, you know, if you can't remember it exactly. I, you know, just if you don't get the right term for this thing, if you know it exists, that's good enough for me. I'm not here to test you on this. And like, especially if we're doing any kind of like live code type stuff, I'm like, I'm not looking to see if you've remembered the semicolon or if there is the brackets are matched or, or whatever. That's unimportant here. Right? Right. I trust that you, with an IDE, you would actually do the right thing and not, and tend to avoid those things anyway, 62 00:21:20.100 --> 00:21:20.880 Yeah Yeah. 63 00:21:20.880 --> 00:21:40.560 Yeah. It's, it's interesting. I think there's two things that go into that. There is the, first of all, as you say, humane, I like the idea of being humane. This, this is another human being who is stressed out and worried, and you know, your job is to say, are they a good fit for our company? Which is unfortunately puts you in a slightly antagonistic position in a way. Yeah. 64 00:21:40.560 --> 00:21:41.360 Yep. 65 00:21:41.360 --> 00:22:07.950 But there's a sort of second point here, which is like the vast majority of people that want interviews, I mean, certainly, and again, my own pos position in the pipeline notwithstanding, but the, the vast majority won't pass the filter at some point along the way. And, and then, you know, the HR department is gonna have to make a, uh, the awkward call to say, thanks very much, but it's not a fit for us or however they do it, I'm sure they're much better than I am at phrasing bad news to people. 66 00:22:07.950 --> 00:22:08.720 yes, 67 00:22:08.720 --> 00:23:02.360 I think it's, it, it's great. Another part of that humanity, the humaneness, I think, uh, is to make sure the person who's been interviewed felt that everything went fairly for them, and that they come away feeling like it was a good conversation. That was okay, actually, even though I didn't get it, maybe it was just a cultural thing. Maybe I didn't quite understand the thing. Maybe the role wasn't a fit for me. Not like, oh my God, those people are awful . You know, nobody wants that to happen on either side. And that's, that's a difficult thing to pull off when you are in this sort of antagonistic situation. You know, you want people to kind of walk away and say, oh, well, maybe next time or that kind of feel or, or something like that, they feel still positive about their experience. Um, and so being interviewing humanely is a great way of, of, uh, uh, allowing that to be well, giving you an opportunity to, to have that experience, um, have the candidate have that kind of an experience. 68 00:23:02.360 --> 00:23:39.800 I mean, I've seen things before. I've, I've, I've sat in other people's interviews, not at this company, but at one a couple ago, uh, where somebody asked a question and then they will sit sort of silently just staring out the candidate. And I felt awkward as the sort of like like the other person in the interview, like learning how to interview. I'm like, give the, give the guy a break. Now I'm probably more likely to step in and say, oh, maybe I can help you a bit early. And that's something that's on me to learn and whatever, but like, it feels extremely, like I'm looking for the cheat, the, the, the one word phrase that is the exact answer for me. And I'm just sort of gonna wait for you to say it. And I, I think that's, that feels too, 69 00:23:39.800 --> 00:24:30.580 Yeah. You know, I, maybe this is just my own personal opinion, but I, the time that you have in interviews with candidates is precious. You're using their time and you're using your time. And the question as we said at the very start of this conversation is, I need to get a sense of what their process is gonna be like in the first few weeks and months of the company, right. As they're coming on, as they're ramping up, as they're getting to a point where they're valuable and useful and, and, and making a positive contribution. What is that gonna be like, and how long is it gonna take? And for me, that precious interview time, I want to simulate that as much as I possibly can. And I certainly don't plan when they start of asking them very one-sided questions and then sitting in silence with my hand on my chin going like, what is the answer to this? Like that, how their first few weeks are going to go, 70 00:24:30.580 --> 00:24:34.160 It, it should be a collaborative thing and you're ready. Right? Yeah. 71 00:24:34.160 --> 00:25:45.320 Right. So I want to try to make that as sort of as realistic as I can, as collaborative as I can, and as close to I can a, as the first few months, and I am trying to do some things to basically like compress that time, you know, compress like weeks of that collaborative time into like an hour by doing a little bit more things that are a little bit more directed and a little bit more pointed just to like, try to get as much information as I can. But the goal remains the same, which is, you know, in those first few weeks and months, what is your trajectory gonna look like? Right. Because if it looks like very positive, then you're probably gonna be a great fit. Um, I don't know. I mean, this is, as we said before, this is just one of those things where you, you're always gonna be able to look back and sort of be like, yeah, you know, it's hard to get this right and I don't think we got it right on this candidate or that candidate. Um, and, and as a result, I, I would hope, I would hope that anyone who has done this for long enough sort of has that understanding in the back of their mind and sort of knows, it's like anyone that would, would tell me like, oh, we have a, we have a perfect interview process and we only get the best of the best. It's like, no, , 72 00:25:45.320 --> 00:25:50.400 It's like someone telling you, telling you that they're a C++ expert. You're like, nobody can say that. 73 00:25:50.400 --> 00:25:50.920 No. 74 00:25:50.920 --> 00:25:54.520 The Dunning-Kruger effect means the, if you not, if you think you're an expert, you're absolutely are not. 75 00:25:54.520 --> 00:25:57.420 You are right here on the Dunning-Kruger curve, right? 76 00:25:57.420 --> 00:25:58.380 . Yeah. Right. Oh boy. 77 00:25:58.380 --> 00:25:59.080 Yeah. So, Um, 78 00:25:59.080 --> 00:26:19.320 That's a, I mean, I've seen some very alternative approaches to hiring, interviewing, and hiring. Um, I think, I can't remember the name of the company, but somebody who we got in at a previous company, somebody we got in as a, I think originally as a penetration tester, you know, like a Like security consultant. 79 00:26:19.320 --> 00:26:20.600 Yes. 80 00:26:20.600 --> 00:27:05.020 Uh, he ran for a long time. Uh, Thomas Taichek, I think his name is, it just came to mind, uh, he ran for a long time some like, uh, newsletter with like, like, here, here's like a sort of, uh, brain teaser style challenge of like, can you hack this thing? And then if you did, you could email him the answer and he would send you the next one in the sort of the, the chain. And if he did it, I believe originally, and I, forgive me if he ever hears this, and I've got it completely wrong, but it's something like this, he did this in originally as just like a, this is a fun way to teach folks about security, engineering and hacking and, uh, you know, penetration testing, all those type of bits and pieces. But then he realized that by the end of it, anyone who'd got to like the 15th one was probably a very good security 81 00:27:05.020 --> 00:28:11.340 Researcher type person. Yes. And maybe he should hire them. Whoever they are. . And so he started doing this, and some of the best employees that they brought on were people who had just been through this, and they were very non-traditional background people. Yeah. I think he said that there was one person who worked as like the IT person for a library in the middle of a rural area in the middle of nowhere. And again, I'm paraphrasing everything from memory. Six, seven years ago, whatever. And it's like, there's no way that person without a degree who was just like fixing the computers in the library would've done, would've gotten a job under the normal circumstances of like the kind of interviews that would, would bring somebody in or even bring them into contact with. To even think that maybe I could apply for a job here. Right. No one gave him the permission to apply for that job. And yet Thomas sort of emailed him and said, you're amazing. Do you want a job? And it worked out brilliantly. And I think that's lovely. We all want the situation where, um, we can widen the talent pool. That we can pull from, and we can make it a level, more level playing field, because like it's the, it's not who you know at all at that point. Yeah. I mean, obviously there's a bit of like, whether you've got access to that. 82 00:28:11.340 --> 00:28:13.220 But Yeah, I love that. 83 00:28:13.220 --> 00:29:22.060 It was an interesting story. I love the idea. And he set up, I believe he set up a company to like essentially, uh, uh, productionize that approach to hiring. And he said literally their, their, they, their final, um, interview process was someone would come in and the HR person would ask them like three questions only. Like, what is your name? Where do you live? Got, got a very, very straightforward, and that was all the interview was, was like, are you a real human being, effectively. And I think that sounds great. And I think it worked for him because so much of it was individualistic. So much of it was, um, uh, uh, consulting work, remote consulting work. I still think there's a bit, and this is unfortunately, this is where now we're into the "Oh, but" there's a bit where you have to work with someone and you have to be able to, Um, interact with them on a personal basis, and you wanna, uh, make sure that you're gonna be productive as a unit of people working together. Perhaps you are gonna be customer facing, in which case, you know, you have to be able to deport yourself with a customer in a way that doesn't cause issues as well. Right. So that obviously slightly takes the edge off of that approach, but I like the idea. Yeah. I like the idea in, in print and principle. 84 00:29:22.060 --> 00:29:50.220 Yeah. I, I love that. And I mean, one of the things that is particularly compelling about that is not only does it work, which is the most important thing, but it has, uh, a nice sort of, um, uh, balance of, of cost. So one of the things that we, and we were just sort of saying this think about a lot is like, all the people that are participating in this interview process also have their day jobs that they actually need to do. And all of the time that we're spending on this takes away from that, which is what we use to, you know, make payroll. 85 00:29:50.220 --> 00:30:05.000 Mean, you think six hours for an individual candidate could easily be taken up in all the things that happen. Maybe even more than that. Uh, yeah. And that's a whole working day for a human being at our company. Per candidate. You can't, that that's not cheap. 86 00:30:05.000 --> 00:31:00.380 Right. Exactly. Right. Um, and, and you know, it's, it's, it's expensive for the candidates too, but like, especially, when you have a lot of influx that sort of, the burden can sometimes disproportionately fall on the company, um, especially because there's like, um, deficiencies of scale there, right? Like the more candidates you have, the more, the harder it gets to organize any one candidate. If you have one candidate a week, you, they're loaded up in your head. You know who they are, they're moving through the pipeline, you're keeping track of them. Um, the more you get, the more those kinds of informal systems start to break down. Um, and so, so the, the thing that I really like about this is that, you know, once he's got those problems, it's really easy to just be like, oh, I got a question from somebody. I'll check to see if their answer's right. I'll give them the next problem. That's like a minute. Right? And, and, and it, it does sort of put a lot of the burden on the candidate to then solve the problem, like it takes time. 87 00:31:00.380 --> 00:31:02.180 Which is another issue. Right? But. 88 00:31:02.180 --> 00:31:22.620 But, but that's sort of like, if you're trying to like, design these systems to be able to handle this, like that's, you want to be able to have those tools in your toolbox where you can say like, I'm gonna take a minute to send you an email and you're gonna work on it for an hour, and then I'm gonna get a lot of signal back from that result, which then I can move you on to more higher touch processes. We're investing more time. Right. Um, before we, which is get to that point. 89 00:31:22.620 --> 00:32:06.620 Which is a little bit, I mean, so now we're, we're going all over the place here. Uh, but like you mentioned, like the leet coding. Uh, type thing that is sort of what that is trying to do there. I mean, in particular, my understanding is that it's the kind of thing that companies might be tempted to do around campus recruiting time where there is a huge, uh, imbalance in power where if your company is well regarded enough to go onto a list somewhere that's on like a GitHub of some college, then maybe someone is just gonna throw their resume at hundreds of companies. Because it's cheap to do that these days. You don't have to print it out and hand it to people. You don't have to send it in the mail, even you just email it. And so you can make a massive list of companies you wanna apply for, and you can be like, not even necessarily very committed to them. 90 00:32:06.620 --> 00:32:48.980 You're just throwing a wide net. And who wouldn't try that? Oh, yeah. I, I'm, I would, yeah. Yeah, yeah. So, but, but that means that as an individual company, even with one that's relatively small and, and, and not necessarily as well known as ours, gets thousands of applications from people that are graduating. Right. And it's not the case that every single one of those people wants Absolutely. To work at Aquatic. Right. Right. That's not necessarily the case there. It's maybe like they, like, I just wanna work in finance or Chicago or wherever. Right. There's just like, you, some search term found you and why the heck not? And then it's kind of incumbent on us to filter them down to the few to say like, well, do we act, is this actually worth proceeding? Or were you just having a laugh? Right. Having a, well, yeah, you're looking for a job, right. But 91 00:32:48.980 --> 00:33:10.720 You're looking for a job and it makes sense. And I mean, certainly for those large companies, like I totally understand why they use leet code and all of these other tools to do this filtering. Like it is kind of inhumane and I hate it, but it's like, I totally understand why they do it. Um, it makes sense for them. I'm not saying that they shouldn't do it. Um, for me the solution is, 92 00:33:10.720 --> 00:33:33.000 But I say it's like, it's, it's a degree of scale compared to say the, the, the, the, the, the process that I was just describing with Thomas Taichek stuff, you know, like it feels a little bit like it's the same in in some ways, you know, you're making, you're pushing the burden back at the candidate to say you have to prove yourself a few times before I invest that little bit more time. And that can feel inhumane. Very definitionally inhumane because it's almost automated. 93 00:33:33.000 --> 00:33:34.020 But don't discount the fact that this feels cool. , 94 00:33:34.020 --> 00:33:35.000 Right? What was that? Sorry? 95 00:33:35.000 --> 00:33:56.040 Don't discount the fact that it just feels cooler. Right? , like in his particular instance, the thing that he was doing was awesome, right? It's a security researcher. I'm giving you this crazy puzzles, you know, you could do a thing where you like, sort of like put some Easter Eggs in places so you can find the email address to the, that's like, don't discount the value of awesome. Right. Like that is just, awesome. 96 00:33:56.040 --> 00:33:58.960 Is a very good point. That is a very good point. 97 00:33:58.960 --> 00:34:02.060 Right. Building out the leetcode thing for Facebook is not awesome. 98 00:34:02.060 --> 00:34:13.080 . No, that is true. Yeah. And these things also, I mean like the, the, the, the solutions get posted and then you're back into the world of like, well, if you know where to look for the solution, then you can jump through the thing anyway. Where, right, 99 00:34:13.080 --> 00:34:13.460 Right, 100 00:34:13.460 --> 00:34:18.220 Right. So it's, I think recruiting is really difficult, 101 00:34:18.220 --> 00:34:20.240 , it's, it is really, really difficult 102 00:34:20.240 --> 00:34:21.580 For on both sides. 103 00:34:21.580 --> 00:34:22.080 Yeah. Right. It is. 104 00:34:22.080 --> 00:34:27.580 I know we, we've we're fortunate to be on this side of the recruitment Yeah. Um, conversation. 105 00:34:27.580 --> 00:34:35.300 Yeah. Right now. But, you know, it's like, we've both been , we've both been in that world where it's like, you know, oh, okay, I guess I'm gonna do some interviews this week. See how that goes. Right. 106 00:34:35.300 --> 00:34:37.200 Yeah. Yeah. That's true. Yeah. 107 00:34:37.200 --> 00:35:04.620 Uh, yeah, that was, that was my, uh, personal anecdote. That was, uh, actually how I wound up at, uh, Dr. W . Prev-prev-co is, I was moving from Texas to Chicago, and I had a whole week blocked off that I took vacation for, uh, to come back to Chicago and interview with like a whole bunch of companies. Um, and, you know, that was a very stressful week. 108 00:35:04.620 --> 00:35:05.380 . I, I bet. 109 00:35:05.380 --> 00:35:11.860 Yeah. This was back in 2010. So obviously all these interviews were in person and they were in offices, and it's very different than the way we do things now, but, uh, 110 00:35:11.860 --> 00:35:51.700 We do things these, yeah, right, right. I know, I haven't really interviewed someone in person physically for a very long time, but Yeah. Two, 2010 also for me at, uh, DRW that was end of 2010. Uh, I was, I had flown to Canada because I was doing something for, oh, I was still working at Google at the time, and I didn't, I was talking to DRW, but I wasn't really sure about it. And I certainly didn't want them to pay for me to fly internationally from the UK where I was living at the time. All the way over to, to, to Chicago and then like, put me up in a hotel, a thing that felt a bit much like I was committing to them. You know, like they were giving me something and then in return I would have to give them something kind of feeling. 111 00:35:51.700 --> 00:36:18.080 And I was not sold on the whole idea, but I was in, uh, Toronto or nearby, uh, kitchen or wherever the heck the, um, Google office was over there for something. And then it was kinda like, well, it's just a little puddle jumper across the lake to get to Chicago, so All right, I'll do it then. And then, yeah, it was a whole day. It was a, a Friday, which I was meeting a friend as far as Google was concerned. And I, I spent chatting to a whole bunch of folks, most of whom now are very good friends with, 112 00:36:18.080 --> 00:36:19.120 Yeah you did meet a bunch of friends that day! 113 00:36:19.120 --> 00:36:24.860 Actually everybody I interviewed with , everyone I interviewed, so Nevin Liber, uh, Oren Miller. 114 00:36:24.860 --> 00:36:25.860 Oh my God. 115 00:36:25.860 --> 00:36:32.130 Uh, all the people that we know and love and we still hang out with at different, uh, uh, were in those interviews. It would, uh, it was, it was a fun time. 116 00:36:32.130 --> 00:36:33.260 Oh my God. That sounds great. 117 00:36:33.260 --> 00:36:42.600 So anyway, uh, that was, yeah, you started it with a personal anecdote. Made me have to say one. I think this is probably a place we should stop talking. 118 00:36:42.600 --> 00:36:50.600 This is a great, I think this is a great way to end it. I think this is a great way to end it. Cool.