 recording. So just tell me when you're done. Hi everyone. The talk I'll be giving is the Humane Technical Interview. So as somebody else said earlier, if this isn't your talk, go ahead and leave. Don't feel bad. I like that. I'm stealing it. So, who am I? So the first time when I, you know, go a little bit about me and I swear it's actually relevant to the talk, unlike, you know, some people. So I work at a DC area startup called Speak Agent and I work there as the CTO or as I like to call it last nerd standing. So a while back we realized we needed to hire somebody because I was the only coder and we only had one front end designer and we were extremely beleaguered. So our CEO comes to us and says, okay, well we need to give them a technical interview. And I'm like, yeah, totally. And then there was this awkward pause and he's like, do we have one? And I was like, oh, no, I'm the last nerd standing. I have to make this. And I was horrified. So I actually got to thinking about why do technical interviews suck because I've always hated technical interviews and I've talked to many, many people about them and many, many people also hate them. I can count on one hand how many people actually enjoy technical interviews. So I started asking myself why do they cause so much strife? And this is not a complete list. This is a non-exhaustive list. But, you know, it's one of the things I kept hearing as I talked to people and interviewed people about these, you know, about these problems. And there was one underlying problem. And that's because we're often not testing what we think for testing when we do technical interviews. Because what do we want? We obviously want a skilled coder because we need to know that they can do the job and we need to know they can do the job quickly. And we also want somebody who works well under pressure because business moves quickly and we always have deadlines and our business is especially under stress all the time. So we need somebody who is grace under pressure. We want somebody who gels with the team. So because we want to all get along and we're one big happy family and we'll work better together if we are all one cohesive Borg-like unit. We also want the best candidate because isn't there always a best candidate? There's one ideal candidate out there and you're going to find them and you know exactly how to find them. And since time is money, we need that hunt to be done quickly because recruiters are expensive and interviews are expensive and having a blank spot is also very expensive because your coders are dying because they need another body in the room. So how do we find these people? One quick warning before the rest of this talk. I pulled a lot from real life experiences, both that I went through and that people I know went through. If you recognize yourself in any of these, I'm not naming any names, but you may be able to recognize something. Know that you're not the first person to do this and there's no hard feelings. I don't want to hurt your feelings. This is just something that is usually happening throughout the industry. Just you made the best story, but I'm not naming any names. So you shouldn't be able to figure it out. Hopefully. Don't Google it. So first off, let's start with a skilled coder. One of the first tests we do is we give them fun coding challenges. I mean, you know, you might ask them to code a game. So you're not a game company, but hey, who doesn't want to code like some cute little game and isn't that way more fun than like reversing a tree or doing anything boring? Or better yet, you might invite them to come to a hackathon and that way you can buy them some pizza, get to know them. They can write some code for you. You can look at the code. It's a great situation. Or you might actually give them some silly code and say, Oh, look, here's a silly code. Look at it. You know, answer some questions. Here's some silly requirements. You can pop in there. Isn't it funny? We're all very funny. We like fun. So what you think you're testing? You think you're testing to see if they know how to code? And you're checking to see if their code is clean, because we always want clean code. And you're also seeing if they are awesome, because you only want awesome people at your company. You also want to know if they like fun because, Hey, we're fun. Everyone likes to have fun. You don't want some, you know, gloomy guts getting in your works. And fun is important at work. But you're actually testing. Do they know the language they're working in? This is legit. Yes, test for this. This is good to know. You want to know that they can code in the language that you are working in. I am big on this, because people often say, Oh, you don't want to see coding challenges. Actually, I do. I think coding challenges are fine. But I think you can very easily make them suck. So let's not do that. Do they like your fun? I think sometimes we think that there's only one kind of fun and we're having it. And we all have the same fun. But believe it or not, there are developers who don't like Monty Python, who, yeah, I know, right? Who don't like Dr. Who or Star Trek or Star Wars, who don't like science fiction or fantasy, who are just very serious people and they don't have the same fun as us and that doesn't make them better or worse coders. The other thing you're testing is can they fake liking your fun? And I've done this because there are things that are popular in the tech community that I don't care about or that I actively dislike. And when I was a younger coder and more vulnerable to backlash, I definitely nodded and pulled up Wikipedia entries so I could talk about something, but I actually did not care about that thing at all. So you're seeing if they can fake liking your fun. What are you missing? Well, you might be missing people who have a hidden background in your product. If you're coding challenges, don't mirror your product, they're not getting a good chance to get down and dirty with your product. This is one of the things I discovered in my company, we mirrored the product. And we found out that one of the people we were interviewing has a background in the product. I was kind of amazed, but he didn't think of it that way. Our product is for ESL. Children who are learning English as a second language. We didn't realize the guy who was interviewing with us went through ESL. So he actually put that in the code and he didn't think to mention it. Because and I'll go over this later, but it came up in the code. And he was like, I'm like, this is amazing. You know how to do some of these things already. But it's not something that he would put on his resume. So what I suggest people do instead, mimic your actual product, you don't have to have them actually hack on your code, but just mimic it. And it's not a bad idea to mimic the most common day to day tasks. Because let's be honest, none of us come into work every day and build a system from scratch. If we did that, it would be terrible. What we do is we come in and we fix bugs. And we add a new like a tiny module. And we might do a little bit of planning. But you know, mimic that instead. So they can also get an idea of what it's like to work at your company. Because you don't want to have all this fun stuff, and then hire them. And then they realize all they're doing is making slight modifications to widgets for the next 20 years. Another thing we do is quizzes. So you might like these are usually pop quizzes, and they're usually verbal. So you're like, Hey, do you know the difference between different sorts? Or what a certain Unix commands do that kind of thing? So what do you think you're testing? Think you're testing basic dev knowledge. And you think you're testing basic knowledge of the language or coding. And you're thinking you're testing of knowledge of more advanced concepts. But let's be honest, be trees and sorts, the vast majority of devs don't touch this once we leave college. I have not sorted anything, except for using the sort command in probably 10 years, like that's been it. That's all I've done. And I do a lot of development. I don't even think I have come within 10 miles of a B tree, or any kind of tree that I've had to actually interact with. So, you know, some people do, but I would put money on it that most people are five or more years out of college just aren't going to remember these things. Also, if you didn't take a traditional path, you may not know some of these things. I teach people that are transitioning from one career to another, and they're skipping the computer science degree. And none of us go over this kind of stuff, we just don't we're like, No, no, no, you need to know how to do things in Python, you need to know how to use, you know, Linux and stuff like that. So they would know commands. But I'm not wasting time like teaching them like, here's a bubble sort. No, they don't need to know that. They're grownups. And you also find out who crammed for the interview. I am in several technical slacks. Each one of them has some kind of interview channel. And in each one, at least once a week, we have somebody coming in saying, I have to cram for my technical interview. What are some good resources? And this is kind of crazy because this is not getting you better developers. If they're just cramming for these kind of BS tests, they're not better than somebody who didn't cram. And the last one, people who have the docs memorized. Some people do, some people don't, most of us open up a Google tab like 20 times a day. Because I know almost everybody in this room has had a point where there's so many tabs open, like with search results, you can't actually see any of the titles anymore. Because this is all of us. Real quick, basically wrote, you know, the bomb blows up because you can't do a valid tar file. I can't do that. My husband, who is a sys admin of like 20 plus years, is like, oh yeah, NASA would be gone if this happened in our servers. So what you're missing, you're missing highly skilled people who don't have computer science degrees. And you're also missing people with poor memory for college. I actually took some computer science classes, though my degree is in psychology, and I don't remember half the crap that I did there. And I got straight A's. Like if you put me in a room, I'd be like, I probably couldn't pass the first exam anymore. So what to do instead? Stick to things you actively use on a weekly basis. It is totally fair. If you use virtual amp every day, make sure they know what virtual amp is. If you actually, actually do have to touch bee trees, and there are companies that do have to touch trees and do have to deal with sorts, you know, very intimately, like, yes, it is totally fair. If you have to do that on a weekly basis, go over it. But if you don't, you're just, you're just, it's like nerd cred, you know, it's meaningless. Sorry to tell anyone that nerd cred is meaningless, but it is. If somebody freezes, and we'll go over this a little bit more as we go on, if somebody freezes, follow up with them later. Because anxiety is real, and this might be what's happening. You might be misinterpreting what you're seeing. And consider not doing them. Like seriously, they're really, really bad. They don't mimic real life. Because developers have what I call keyboard, like their brain is in their fingertips. So you only know how to do something if you are actually touching the keyboard. So if you've ever tried to walk somebody through doing something technical over the phone, like a relative, and you haven't like wanted to throw the phone within the first like 30 seconds, you know what this is. It's really hard to explain things. So the text test, whiteboard challenges. Oh boy. I love these. So for anyone that doesn't know, by the way, this is when you put developer that you're interviewing in front of a whiteboard and say, write out some pseudocode to do blah or design this or draw things like random, random coding things. So what do we think we're testing when we do this to people? Well, we think we're testing how people think on the fly. Like they're not at a keyboard. They can't Google anything. What's their brain working like? And we also think we're testing, again, knowledge of computer science concepts. But what you're really testing is who really who can speak out loud easily. Now, I'm a teacher and not like every day, but I teach for fun and occasional profit. And I am good at thinking aloud. I am not good at thinking loud because I'm a developer. I'm good at thinking aloud because I've taught so many classes. I can do this really easily. It does not bug me often. I'm also a weirdo who talks to herself all the time. So I'm just used to talking a lot and thinking aloud. So it's a skill. It's a very different skill has nothing to do with how well you code. I assure you, I am not a better developer because I'm a weirdo who talks to myself. You're also testing who is a visual thinker. We're not all visual thinkers. Some people just can't do it. They're not better coders. They're not worse coders. They're just people who think a little differently and have maybe trouble drawing things out. And again, people who have crammed for the interview because there are people who will go on YouTube and watch people do the common whiteboard examples, memorize them and wait for the question. They are not better coders at the end of the day because those YouTube videos are just like how to get past this stupid pseudo Google interview question, which by the way, Google's the one that started this terrible, terrible thing. Even they're like, oh, we're not doing this anymore. Like even they're stopping it. So maybe you don't do them. What to do instead? Just don't, just don't. Like, you know, this is, they're hard. Like, first of all, if you have a remote job, you've got to pull people into an office. Everybody gets stared at. They're dumb. Don't do them. Like just don't. Hidden rubrics. What do I mean by hidden rubrics? Well, it might be actually a sheet that you have where everyone scores the developer because, oh, that's really, you know, it's, it's neutral. We all know what we're grading on. You know, you might put in like hidden tricks, like code, like hidden bugs or instructions and wait to see what people do, that kind of thing. So you think you're testing how good a coder is. Again, like everything we're doing is just to see are they a good coder. And you're seeing, you're trying to see if they pass a sniff test. Like you're trying to see, you know, oh, are they doing anything that might make our spidey sense go off or something? And you want to find somebody who meets the criteria for a job and you don't want to be biased. You're worried about that. You're worried that you're biased, that your other people may be biased. You want to be fair. So what you're actually testing is who got lucky that day. Because sometimes you have somebody who is just going to score them higher. Maybe they like them more. Maybe they're just in a good mood. Because if somebody's in a good mood, they are much more, like, likely to score you higher. Maybe they just had lunch. Because by the way, don't ever, ever sign up for an interview before lunch because people get the hangries and the hangries are real and they actually have real world consequences. So, you know, that person just got lucky with their time slot. And this one, who got insider info? So this happened to me. I was interviewing for a job and the recruiter really wanted me in that spot. Just like, you're perfect. I'm going to get you in there. Do you have 10 minutes to talk before your coding challenge? And I was like, yeah. And she called me. She's like, yes, okay, these are not in the instructions, but you need to do A, B, and C. I was like, what? And she's like, yeah, if you don't do A, B, and C, they won't hire you. It's a hard requirement. Like that was one of their hidden rubrics. And I'm like, but okay. And I did A, B, and C. And I'm like, yeah, it's amazing. Nobody has done all three of those yet. And I'm like, I think I know why. And I've had other people that have had that happen to them. And it's, you know, they have a friend that gives them the tips. So all you're checking to see is if somebody actually got insider info. And also, who can get by your hidden biases? There is a great book called Weapons of Math Destruction. And it has a really cute title. And then you read it and you want to die. Because it's so depressing. Because it's about big data. And it is about how big data can make you feel like you're being fair, but you're actually being worse. Like you would have been better off going with your racist gut than if you had just instead of this rubric, because the rubric was worst. Because we all have hidden biases. And those biases make it in to the rubrics. When you may have a bias against like people who didn't complete college, who didn't go to college, who don't like the right, who haven't had the same paths, who don't use the same text editor, whatever. Like you have biases. So what to do instead? Be clear. Like tell them what the rubric is if one of your requirements is that you have to have mock tests. Like tell them. It's really weird. But just tell them. In order to pass our rubric, you need to do A, B, C, and D because this is how jobs work. Like in order for you to get paid this week, you need to do A, B, C, and D. You don't come into work like, oh, you didn't pass our super secret test. You don't get a paycheck. Like that doesn't happen. And maybe cut it down a bit. You know, you don't need a huge long rubric to be fair. That's actually not fair. The more things you have on it, the more likely you're going to be end up screwing somebody over unintentionally that didn't deserve it. So cut it down a bit. Because you would be surprised. There's still people, if you have five items on there, there's still people who won't pass it. You know, by the way, the code you submit has to run. I had people that submitted. I'm like, it doesn't, it doesn't run. Yeah, there was an error. And I'm like, okay, well, there we go. Like they already, you're not going to get like 100% pass rate anyway. The next one, gelling with the team. This is not technically something you do during a technical interview, but it is so totally something you do during the technical interview. Because what did the developers do after the interview? They're like, oh, do we want to work with this person? So it's part of it. You have to think about that. The first test is culture questions. Yeah, if you thought the previous section was a little rough, buckle up. So what does this look like? They often pop up as questions that are not about coding. So what do they do in their spare time? What kind of media do they like? And what's their Hogwarts house? And that kind of stuff. Nerd crap. So what we think we're testing, will this person fit well with our team? Because we want somebody who will fit well and it'll be seamless and wonderful. And will we have something to talk about? Because we are one big happy family and you have to fit within the collective or everything will be terrible. So what we're actually testing who happens to have the same hobbies? And not everyone rock climbs. Not even people who say they rock climb. Not everyone drinks. Not everyone goes to music shows. Not everyone does hackathons. And you're also testing who has a similar lifestyle as you. Because you know you like rock climbing. Okay, fine. But you know you're probably going to over focus on that and start pre selecting for people who you could see inviting out or for a beer or because you know that you like the same band. And you spent 40 minutes in the hallway talking about how awesome their last show was. And oh my God, but did you see this other one in Portland? You see where I'm going with this. So the people you might be missing. People who have different lifestyles. It's amazing. We do exist. Because not everybody has the same lifestyle. Some people have kids or aging parents. Some people have demanding pets. There are people who just don't like to go out. And that's that's fine. Like, that's okay. People who stay at home and sit, you know, so and knit and watch YouTube all weekend are, you know, that's me. That's what I do. That's my exciting life. I don't do much else other than that. That does not make me a better or worse cutter than anyone else. And people who don't care to share because let's be honest. We're a little judgy. You know, we get really judgmental about what people like to do. You know, for years, I actually hesitated to share that I was a crafter because people cut really weird and judgy about the fact that I can make sweaters. And like, everyone needs sweaters. Like, you know, but it's really weird. And you also sometimes don't know what kind of jerk you're going to encounter. So you just don't want to share. Another one is Spoonies. For those who don't know, Spoonie is a term that's used to refer to anybody who has some kind of condition that leaves them with severely limited amounts of energy. So people who are Spoonies, and you can look, the thing is at the bottom, I'm not going to relay the whole story. It's a long story, but they have to spend their reserves very, very carefully. So everyone I know who's a Spoonie who works full time, their weekend is going home and recuperating. Their evenings are going home and recuperating because they're trying to bring as much energy as they can back into the office on Monday. And that's why they may not say that they do all these cool things because they're sleeping and they're resting and they're doing self care. That's why, like not because they're bad coders, rock climbing has nothing to do with coding. So what to do instead, stop putting people on the spot. Like normally I advise people like say, oh, tell us about yourself, what they care to share, they will share, don't start quizzing about weekends, don't do the friendly chitchat. First of all, it's also keeping them in the office for another like freaking hour. And interviews are already long enough and they're going to be able to let them choose what they share. Because when an interviewer asks me what I do on the weekend, I sense a trap. And I immediately start overthinking it. And I'm like, okay, okay, okay, let me do a read on this guy. Okay, I need to talk about, like, you know, maybe movies, he looks like a movie buff. Okay, I'll go with movies. And I'll go with hackathons, maybe sometimes. And I will try to do a read on the person and tell them what they want to hear. They can find out after the interview that I actually sell all weekend. And stop weighing this when hiring, even if it comes up organically, even if you just end up talking in the hallway for 40 minutes about your favorite band. You do not want to penalize candidate B, because candidate A does something you like. That's really horrible. Don't do that. Like, really try to step back, be aware of your biases. And yes, they're cool person, you want to hang out. But you do have to make a choice based on like, I don't know their code and stuff. And also, what I do on my weekend is none of your business. Like, keep that in mind. It is absolutely none of your business. If no matter what you're doing. So anytime you start thinking and weighing that, just have that in the back of your mind. And so I made a nice one. The next is pseudoscience. One thing I like to remind people when they're writing interviews is that they are not a social scientist. And I said this as somebody with a background in psychology. I'm not even a social scientist. I can just kind of fake being one sometimes. And I wouldn't do these things. But we really like to pretend like we're social scientists when we do interviews. So they'll probably come up as like little subtests, little traps, you know, you're watching them really carefully, that kind of thing. So we think we're testing hidden attributes. Because we want to see if we can suss out any of those things that don't come out on their own. Like, you put a puzzle in the waiting room. And then you have the receptionist watch to see if they touch the puzzle, because that indicates that they're curious. Kind of like those like, like on your first birthday in, you know, like China, they put a bunch of things in front of the baby. And that's their profession. Like that's kind of what we're going for here. If you touch the puzzle, then you're inquisitive. If you touch like the magazines, you know, you like media or something. And I've actually seen the puzzle thing. It's kind of BS. Because I hate those puzzles. You might also try to be suss out deception. So you're watching them very carefully. You're watching their body language. Do they look left or right when they answer a question? Because, hey, we've all watched the mentalist. And that's like super easy. If they look left, they're lying. By the way, it doesn't actually work. The mentalist is a fictional person. Yeah. And sometimes we try to suss out a true person by eliciting what we think of as a real reaction. So I don't know if any y'all have grandmas that watched like detective shows when you were kids, but oh, I'm sorry. Just yes. Yes. Just one more thing. Okay, so for anyone who's not understanding why this side of the room is nerding out right now. This is Colombo. And he is a detective. And one of the things he would do is he would like to ask a bunch of questions to the perp. And he's like, all right, all right, I'm done with my questions. I'm done. You can go and the person would relax and he turned around. Just one more thing. And he would ask some question and it was always like, you know, the person would like show a reaction. Like, you know, why were you wearing a red bow tie earlier, but not now and by the look on his face, you could tell that he killed like whoever whatever. Colombo is also fictional. Just so just so you know. I had this done to me in an interview. And I watched it happen to someone else. The interviewer asked me when it was at the end of the interview, like I was literally getting my purse, like I was packing together my purse and my resumes. And he's like, Oh, just one quick question before you go. What are your three favorite movies? My first thought was, well, I can't actually say what my first thought was because we have we have a code of conduct. But it rhymes with duck you. And but I he's like, What's your favorite movies? Now he thought that by doing this, he could get a read on me. So he was watching how I was reacting. There's one problem. I actually don't like movies. Like, they're fine. I'll watch them on flights and stuff. But I never seek out movies. They're just I don't like them. But I have learned you never ever tell a movie buff this. I've been on that date. It's a bad scene. They always have a movie they want to show you that'll make you love movies. Oh, no. It's like a dungeon and they don't let you leave until you tell them that you really like like burns that burns the guy and it's terrible. So anyway, I actually pulled to out of the air that I'm like, what do movie buffs like? Okay, I'll say these two because I think I know the Wikipedia entry on them. And I'm a childhood favorite. Like, it was a total lie. Like, but I really didn't want to end the interview on that note where I'm going to say like, yeah, movies suck. His read on me. I asked him later because I worked there. And I later on I was like, what was your read on me? And that he's like, oh, you know, he had this interpretation for my movies. He's like, oh, yeah, I could tell that you really like like, you know, dark, arty movies and things like that. But that you also that you know that I'm like, dude, that was total BS. I don't remember two of the movies. You know, and it was wrong and I actually laughed. But it's still like he thought he had a read on me. And he didn't. And that was bad because he was had garbage information. So what to do and you're not actually testing anything because you're not trained. Even social scientists don't do this. If you read actual studies where they're studying like how people like, you know, react to a thing, even they there's a big section in psych papers where it's like where we may have screwed up. And and and those are usually real longer, like, yeah, maybe this doesn't work every time. So that's something you've got to keep in mind. You are not a social scientist. Don't do it. Because what you're missing is anyone who doesn't pass your test. And your test was garbage. So you're rejecting them for a garbage reason. You also might be missing out on those people who caught it, who realized that you were testing them, who got peeved about it. And we're like, I'm not going to work there. Because I'm not going to deal with this kind of crap. If this is how they are, a good candidate just walked away. And you also spent a lot of money getting them in the room. So don't do it. It's not worth the risk. So what to do instead? Just like, don't, like really? Don't do it is I'm not saying don't screen for jerks. If somebody comes in and they're a total jerk, and you're like, wow, this person is a jerk, don't hire them. It's that easy. It's great. You don't need a secret test to like suss out the jerk. So another thing we test for is that they work well under pressure, which always kind of makes me laugh because everyone, everyone is like our company. We are always under such pressure. We are a high pressure environment. There is there's deadlines like y'all invented deadlines. And I assure you deadlines have been around for a while. And I've actually never met anybody who's like, yeah, our work is chill. We never do anything, but we get paid well for it. I you know, if you if you do work there. Anyway, first test, you mess with people. You make artificially stressful situations like inserting a weird bug or giving them a system that doesn't work, or you have an alarm go off or you're just extra jerky. There's a Oh, who's seen Charlie in the chocolate factory? Okay, or one. Okay, there's a guy and he's the evil guy. And it turns out he was like a hired jerk, but he's actually nice. And it works in the movie, but also fiction. Don't do it. So what do you think you're testing? You think you're testing grace under pressure, which is important. Developers, there's not a lot of leeway to like, freak out. We do freak out. But you know, we kind of need to have a little bit of an ability to have grace under pressure and ability to work in your high pressure environment, which went over, you're not special. What you're actually testing for is the ability for a person to manage their stress in that situation. And let me explain what that means. When people aren't a situation that is familiar to them, they are able to act under pressure. If they're in a situation that is not familiar to them, they're not going to be as good. So I am a boss at putting on fires in my own kitchen. If you put me in somebody else's kitchen, I'm like, well, I guess everything's kind of burned down today. Excellent. And literal kitchens, like other people's kitchens, I can't even deal with where you put things. And, you know, but if you take a professional chef that moves around kitchens a lot, okay, they are trained for that specific thing. They're trained to be bounced around. I am not, I'm not good at that. What you might be missing are people with anxiety issues, people who have anxiety issues. When they're in a familiar situation, they can put together tools and self care, and they can center themselves if they're familiar with the situation. If they're not familiar with it, they have many fewer tools. So one of the things I recommend, go watch one of Ed Finkeler's talks. He talks about this quite a bit. And they're excellent talks. And he gives numbers and stuff like that, which I'm not going to give here. But go check that out because there's a lot of people with anxiety. Either we're born with it and we enter into the field, or it comes up later, whether biologically or because we live in a traumatizing field, one or the other. And people who catch on, people sometimes catch on that you've made this artificially hard that you put that bug in a system on purpose. Some people highly qualified candidates will walk away that they're like, Oh, if this is how you are, goodbye. So like maybe just don't do these because I'm going to tell you something right now below your mind. Interviews are already stressful. Everyone is already stressed. You are seeing if they have grace under pressure by the fact that they're not crying at the table, talking about their resume, like anybody who's able to sit here, they're doing good. So another test is on the spot thinking, which is goes along with the whole like whiteboard stuff, but you also may be like stopping them in the middle of the stuff, something, asking them to explain something. Basically, if you're ever hawkishly watching someone and making them answer something right there, this is on the spot thinking. So we're looking for ingrained knowledge. This is a very specific skill. Again, I'm a teacher. I am really good at explaining things on the spot. I have a lot of answers rehearsed, but it's not a it does nothing to do with my coding. Like that did not make me a better coder. And you're also testing who is an anxious. If I am anxious during an interview, I lose the ability to explain things. I had that happen where I couldn't explain how to write a Django view. I write Django views all the frickin time. It's my job. And I sat there and I blanked. I completely blanked. I'm sure the guy is like, how has this woman not been eaten by wolves? But that was it. I was anxious because he was staring at me and he had these dead soulless eyes. And I'm just like, no, get me out of here. So, you know, again, you're missing people with anxiety issues. And they once you get them in the office, they're fine. Like once they have a setup, once they're familiar, once they don't have somebody staring at them with their soulless dead eyes, they're fine. Also, people who have finger memory, because as I said before, developers tend to work best when we're staring at a screen and typing and our mouths aren't moving. That's just how we work. It makes sense. Most of us do not code with talking. I've only met one that did that. So what to do instead, make it natural, put them in a natural environment of an environment they're comfortable in, let them use their own computer, let them use a familiar setup, and just, you know, let them work as they normally would work, because that's where you're going to get the best data. We want good data on these people, not crap data. And if you must allow for followups because anxiety does happen, you may want to follow up the next day because what people do when I was at the view thing and I'm like, oh my God, what even is Django? Like I was having a moment, I hung up the phone and was like, what the hell is wrong with me? I know what a view is. And you know, I did end up following up with someone the next day where I explained the answer. And some people are scared like, what if they Google it? It's like, we all Google stuff. Like that's our job is actually to Google stuff all day. And chances are you're going to be able to hear whether they Googled it. I assure you, if somebody is just reading off a Google response, you'll know versus somebody who actually looked into it and actually knows it and just remembered it. So last one, the best candidate for the job. So this is one of those things. We always think there's an ideal candidate. We always do. But I'll let you in a secret. There always, there isn't always an ideal candidate. Sometimes you have a bunch of people that actually would do really well, but in different ways. So the first test we do is staunch requirements. We are going to make it like they have to be an expert in everything we do. We want at least 10 years of development experience, an advanced degree. And you know, they need to like have all these like wonderful things and they need to know how to use the entire stack. So we think we're testing, you know, do they have the expertise for the job? We only want people with the right expertise. We're actually testing who can check boxes. I know a lot of crap developers who can totally check all the boxes because on paper they look perfect. We're also testing who can tweak resumes. This is something you can actually pay for this. I've had people pay for it. Sometimes you're married to the person who can do it. But yeah, like this is a thing. My father could tweak like anybody's resume to get anybody's job. Like he was really good. It was his job. So what you might be missing experience you didn't know you needed that it comes in every once one, you're like, wait, you know how to do that? Oh my God, we could totally use that. And if you pre filter on that, like if you pre filter somebody who has experience making Android iOS apps, and you remember it's like, oh my God, yeah, somebody upstairs said they wanted like this kind of developer. And yeah, they don't know, you know, compass and they don't necessarily they don't have a strong background in like, you know, as strong a background at Python as we want. But this is worth bringing on. Let's bring this person on. You're also missing under qualified people who aren't like if you are really, really honed in like you're going to get that computer science degree, you're missing people who just don't have a computer science degree, but are still solid developers. And people who just don't check the box is quite right. There are people that will have experience in CSS. But you asked for compass and they didn't like say that they could do compass so they don't check the box a lot of newer developers do this. One of the difference between a senior, senior developer and a junior developer is that senior developers walk into a room with like a how hard can it be? Sure, I could probably figure that out. Just how I got my current job, which is how how I had to learn Angular in a month. So, but that's proper, that's development. So what to do instead, match it down your qualifications. You want those people you want people to pass over a bar. You don't need people to vault over the bar. You can set that bar later. But you want a lot of good people to come in. And you want to look at the whole person, you know, just don't focus on these tiny little attributes focus on everything they're bringing to the table. So running a little bit late on time. So I'm going to start blasting through. So saving time and money while searching, it's expensive. So you might pre screened through a third party, like a recruiter, who loves recruiters, everyone, right? So you just want to see if they meet basic qualifications. But what you're actually testing is who can pass by the service and not all the services are doing this correctly. Also, people the services didn't put off, meaning people where the service wasn't such a jerk to them, that they didn't say never call me again. So what you might be missing people who just failed their weird tests, which by the way, I've been failed by them. And I'm like, actually, no, I was right. And they're like, well, technically, I've Googled this. And this is the Google says this is the right answer. This happens. People, the recruiter rejected for non tech reasons. And this does happen where the recruiter just does not like the sound of your voice. Because recruiters can be just as sexist, racist, ageist, or any as you can come up with as anyone else. Also, people who reject the recruiter, I had a recruiter that was a total jerk to me. And I said, if anybody calls him calls me from your company ever again, I swear to God, the tweet storm that will land on your front doorstep will be ugly. What to do instead, consider doing yourself. I understand it's a lot of people, but maybe just considered not using a recruiter. If you and another thing you can do is start with the technical challenge. That is a good way. Like if somebody says, hey, I'm interested in your company, you're like, cool. Here is a coding challenge. Do this. We'll talk. People will self select. If they look at it and they're like, I don't know if I think I don't think I know how to do this, they'll walk away. Congratulations, you've saved time. If somebody says, like, oh, I don't actually know Django, or I don't want to work in Django. They'll walk away. That's fine. Some people will just, you know, decide not to do it. And if you do use a recruiter, vet them often. Call people who have been interviewed through them and say, what was it like? What happened? You know, even if they don't get a job, what was it like? Can we get a few hours a year? Can we get an hour of your time? We'll pay you. That kind of stuff. Like make sure you're checking to make sure they didn't like have somebody else do it for them. Aggressively filtering resumes. This is like having this not requirements. Again, expertise. But you're only testing for surface qualifications. And you're only testing for resume writing skills, which by the way, you can purchase people to do this for you. So you might be missing innate biases, like, you know, who went to college, hidden qualifications, people that have those secret hidden skills you didn't know about, and good kind of candidates with less than stellar resumes. So ratchet down again, the qualifications, chill out and shorten up the process. Because one of the things is we have these long processes because we have all these little tests in there and everybody's got to be in the room. Everybody, you know, we've got to get everyone in and everyone's to talk to each, you know, talk to the person. And we get garbage data anyway. So why not just focus on getting some good data, a smaller amount of data in taking up less of everybody's time. So what do we do? I would say make the coding challenge relevant and short. And I would also put it at the beginning, you know, let people see the code, have a quick conversation and see if there's interest. Don't spend four hours of their time before you've even seen the code. And again, do the challenge first. Shorten the interview. You do not learn anything new about the person in the fourth hour than you did in the first. I swear the only thing you've learned is who can smile the longest and keep their cool and not go to the bathroom because you never want to go as to go to the bathroom. Trust your team. If Joe can't be there, don't bring the person back in because Joe needs to talk to them, reschedule or say Joe, trust us, we're not going to hire a serial killer. And Robert, the interviewee is the interviewer. They are looking at you as much as you are looking at them. So you want to put your, you want to put a transparent, you know, everything on the table, everything out there because you want them to love you as much as you want to love them. And with the three minutes we have left, do we have any questions? I've seen people ask that, like interviewee is asking to see his question was have you seen an interviewee ask to see the inter the company's code? I have seen people ask that. I've never like in some companies will do it. A lot of companies, if you're an open source company have like a repo that they can go look at and look at the code. I did have one of my and actually a bunch of our stuff is public. One of the people I interviewed went and looked at the code. And I was like, don't judge me. I didn't I didn't know angular very well back then. And he was very nice. He's like, I could fix this. And it was great. But yeah, I've seen that and you may offer them the code where it's like, yes, look at us judge us. That might be a great way to pre screen if somebody doesn't want to deal with the hot mess. Yes. It is as being a woman who's been interviewed. I usually counsel guys to be especially sensitive to the anxiety thing because there's I'm like it is no matter how nice a guy you are, no matter how many Pilates t shirts you have bought, it is a fraught situation. It is a heavy situation. Try to put everyone at ease. And as a woman going in nice lipstick or sorry, that's usually what I do. I don't know what everybody else does. But I'm like, I'm going to put on my screw you lipstick. And it's hard, you know, and also, but with women, I usually say find a community of women and talk to each other and share war stories. Like that's one of the things that really got me through when I had a really bad interview. And I was crying afterwards. I went to one of my fellows and my fellow Pilates. I'm like, I used to the worst and they're like, Oh, oh no, I can beat that. And we all went through our war stories. And they also helped tell me it's like honey, you're not crazy. Good Lord. Why did they do it that way? So find a community, find people you can share with. And it can take the edge off of dealing with interviews. That's good for everyone, but especially women. Yeah. And that's when they're wrong. The idea is that the reason you people coming back over and over again is to see who's really determined to get there. And what you've actually tested for is who's unemployed or who has a job where they don't say, Oh, wait, why are you wearing a suit again? Because I had jobs where I'm like, I can come in for literally one interview because if they think I'm interviewing anywhere, I will get fired. Like, so please do not like make me come in for more than one. So it's like, no, that's what you're not testing for people who are determined. They either have no job or they have really flexible jobs or they don't care. So yes. So searching. I and there's lots of ones where I say, Oh, just hide the names and oh, do this. And I'm like, no, I think the best way you can confront your biases is by like, actually confronting them. Reading books about biases that there's, I wish I can remember, I'm really bad with book names, by the way, I only remember the one book because I prepped myself. But there's lots of books about biases and the implicit biases read up on every kind of like social justice thing you can. Because that's one of the ways you will force your gut to stop reacting in a certain way. So I mean, that's the only way the only answer is not to hide everything. It's to put it in front. I know I'm just holding a zero at me. Sorry. I can I take one more? Don't. Okay, one more. Okay. No, your gut is racist and sexist and agist. Thank you, everyone.