Episode 27

Alexey Dubrovin on The Role of AI in Hiring: Balancing Technology and Human Elements

TIM: Well, Alexey, welcome to the Alooba Objective Hiring Show.

ALEXEY: Hi team, thank you so much for having me. I'm very excited.

TIM: Am I? And what better place to start if we're talking about hiring data than AI? Because that is the ultimate in past terms, and everyone's talking about AI in almost every context you could imagine in their day-to-day lives, in society, in the media, in work, and what I'd love to start by talking about is AI in hiring. and have you used AI so far in hiring? Have you seen candidates use AI? And perhaps we can then delve into a few other areas after that.

ALEXEY: Yes, that's definitely your right hot topic everywhere, and what we're trying to do in Carfax Europe is definitely increase the adoption of AI in all areas of the business, be it customer support, software development, or everywhere, although I must say HR hiring processes and recruiting It's a tough one, and there are multiple reasons for that. One is we are based in Germany, Europe, which has tough data protection laws, and as recruiting has to do a lot with personal data, you deal with real people with real names. They often leave their addresses, phone numbers, and everything else on their CVs. and I think what we are, we're obviously always evaluating how we can do it and what works and what not. Can we use some AI language models in a protected environment and do something in that area? But that's obviously the very first concern when you talk, particularly in Germany, about using AI or any sort of data involving personal information. Yeah, it's a tough one, so as of now we don't use it, but I'm very curious and excited to find out ways how AI can help us in that.

TIM: and that's on your side as the hiring company. What about on the candidate side? Have you seen candidates use AI in hiring? I'm sure you have, and if so, where in the hiring process do you feel like they're currently using it?

ALEXEY: I actually have a hot take, I think, on this one. I do see it in between, but I wish I saw it more often, to be honest, because one of the things that is one of my kind of annoyances about candidates CVs and screening process is that you post a job ad, you say very clearly what you are looking for, and you get a pretty high percentage of people applying that either didn't read the job ad at all or they think that it was not worth it to include or highlight somehow their experiences and skills that we are looking for in their CV. And this is so easy with AI nowadays. You can literally upload a job description, upload your CV, and tell a language model to just tell you if you fit or not, tell you what you should add, and tell you what you should highlight in that, so if I can give any advice to potential candidates, just use it more often because it helps, and we'll find ways how to combat it. If we do see that somehow out of 50 CVs that we receive in a day, like 49 fit pretty damn perfectly to this job, then we'll probably start asking questions and start thinking about it, but that's like, from my perspective, a good problem to have. Yeah, can we please have more of that? because then, yeah, I think we'll be living in a more exciting world.

TIM: That certainly is a hot take and very interesting, and the last scenario you outlined, which is we've got 50 applications 49 of them seem perfect. A story or a narrative I've heard quite a lot in the last two weeks where now some companies have said, Yeah, I like it. I'm getting inundated with 200 applications. They all seem to have a very similar style and approach to them, and they all seem good, quote unquote, so I feel like a nuance in the recommendation to candidates might be that if you are a good fit for a job and your CV just doesn't show that, a no-brainer is to use AI to do that. I agree; if you have no skills or experience at all and you use AI to make up a bunch of lies and apply it with a perfect-looking CV, that is bullshit. That is probably not the hack that you think it is. it Probably not, and then I think since there is so far, at least for now, there is always a human behind the process behind the screen since, again, like we prefer not to use it as of now for data protection reasons, it will be visible. I think if you see a CV that is from top to bottom completely AI-generated,

ALEXEY: It's going to be obvious when you look at a good number of CVs and talk to people in the past; you would see it, I think, yeah.

TIM: That's another really good tip there for candidates. Because if you're a hiring manager, yes, you've seen hundreds or thousands, actually, of CVs in your life, and if you're currently hiring, you will have gone through hundreds of applications; you have gone through many tests, and you get an intuitive sense of what is human or not human-written content. and if you're a candidate and, as part of your application, you have submitted your own written content, which is imperfect and is written like you write it, and then you've added a bunch of ChatGPT stuff next to it, as someone evaluating two side by side, it is very obvious.

ALEXEY: Yes, that's absolutely correct, so I think it's similar to how cheating in school works. If you are smart about it, you won't get caught, so I think if you are smart about it, I absolutely advise, first of all, not to cheat in school; second of all, to use AI when you apply for jobs but just not copy and paste stuff. Okay, ask for advice. Ask for things that you need to change. Imagine I'm a recruitment manager, and this is my background. I don't know. Imagine I'm Alexei, a data director in company VC. Pull my LinkedIn profile, like copy-paste it there, whatever, and say what would Alexa say about my CV? Actually, you'd be surprised that AI is quite capable. English models nowadays are quite capable of giving you good advice, and then you do your work implementing those advices yourself, not just copy-pasting, and if you do that, I think the quality of CV will definitely increase.

TIM: That is a great suggestion. I never heard that before. If you know the hiring manager or maybe the talent acquisition person initially, who's going to review a CP? Using their public information to then give you suggestions is a great hack.

ALEXEY: At least try not to. I don't have that many details on my LinkedIn profile that are public, but some people do, and if it's public, capitalize on it.

TIM: 100 That's a great suggestion. What about just philosophically about the use of AI, like you mentioned in passing? If you're going to cheat, well, is using AI cheating, or is it just now something that is just like it's a tool that everyone should leverage?

ALEXEY: I view it as a tool. I don't view it as cheating, but then again, it depends. This is a tool that can replace your thoughts completely with some thoughts, and then it's something just stupid. Why do I need to hire you? I can just ask AI everything I want to ask you and do your job for you. no I post a position because I want to hire a real person and then just use it as a tool to give you advice. I've seen you even like proofread your text; maybe you missed a couple of commas; maybe your sentences can be restructured to be more simple; maybe something is not understandable. I'm not a native speaker. I sometimes make, and sometimes I often make, a lot of mistakes in English when I write or when I speak, and that's such an obvious thing. Scan my CV and tell me what sentences I can make easier to understand. Is the formatting good, or what advice can you give me? This kind of stuff, I think that it's not cheating. It's just you can get all this information online, although if you read through blogs and listen to podcasts, it might take a while until you get a comprehensive picture, and you are able to implement everything with AI. You can take a couple of minutes; you have bullet points of what you need to do.

TIM: What about is there any bit of a hiring process where you feel like using AI, even if it's in a considered way like you've just outlined? Is there any process or any stage that is still off limits? So I'll give you an example. We were recently hiring some people into our team, and we had an application and assessment process. and one of our questions in that process was, Imagine it's day one for you at Alooba. What are the three things you would like from us to give you the best chance of success? So this question was very personal and specific. I just wanted to know exactly what this person needed so I could give that to them. I was amazed by how many candidates used ChatGPT for that, and I thought, I don't care what a large language model thinks about this arbitrarily; I care what you think. don't

ALEXEY: No, but that's a completely fair question because it then goes into the interview process, and you are completely right. You know what? Please don't use it. You can use AI before the interview to again help you prepare and ask what questions you should ask. Here is a company that I'm applying for. Can you look up information? What questions should I ask? What should I talk about? Not talk about myself? Highlight not highlight? Use it as like a sparring partner to prepare for the interview, but on the interview itself, no, please be genuine, please be honest, please be a human being, because we are hiring human beings. We are not hiring robots, at least for now. I think when we do, and/or particular AI models will have a completely different hiring process if we even need a hiring process. No, it's like a marketplace, right? You just go click whatever you can get out there, so yeah, no, I think that's a no-go, which I think if it becomes obvious in the interviews that You are really just saying what AI tells you to say. Huge red flag.

TIM: What we're hearing quite a lot at the moment, I think, is because of maybe suppressed market conditions and AI, and there's then this high volume of applications. Candidates are also seeing this high volume of applications, probably getting worried, and probably then applying for more roles, and it's creating this like vicious loop. And that then means candidates per application are more likely to be getting rejected because there are more applications, so you're getting rejected more often. This sometimes can breed a little bit of distrust in candidates, a little bit of woe is me, I feel like the world's against me. Now sometimes I feel like hiring processes are a bit unfair. I don't think the best candidate always necessarily, gets hired What about in those early screening stages where you're applying with a CV and going through a job ad? Are there any ways that we can make the process a little bit fairer, a little bit more objective, so that the quote-unquote best candidate is more likely to get through?

ALEXEY: excellent question I think a very complicated one at the same time, since I agree that a process can be unfair, and we are all people evaluating other people. We, by our nature, cannot be 100 percent objective. That's why we'll also involve other people in the interview process. It's usually not just one person interviewing, and you usually have multiple rounds with multiple people. We're trying to be as objective as possible, and I remember before I was in a position where I hired people, when I was mostly applying myself, when I started at the university, I heard online and somewhere that when you apply for a job, your CV is looked at, and at most, I don't know, 20 seconds, 30 seconds. I couldn't believe it. It was like, Yeah, but how come I spend so much time preparing it? I look for a job, I apply, I send it out. How come a person can be so inconsiderate that they just look at it, spend a little time looking at what I have prepared, and now I get it: now we do the same. one of the team leads in our company, Giuseppe, and his team are now in the hiring process, and she put it recently in a very funny way because I think it is actually like I can relate to it; I can see how that's exactly what is happening. He said that the face when they screen CVs, they call it as a team. Tinder face because it's exactly like that: you open a CV, and you pretty much almost immediately scan it, and you swipe right or left, and that's, imagine as a candidate, if you ever were on Tinder, that's exactly the time you look at maybe the profile picture and the amount of time you spent to decide to swipe right or left. That's exactly the amount of time the hiring manager of the team decides it's a swipe right or left, but that's very important to keep in mind, right? Because when you know it, you can do some things that might help you, I think. Just imagine what some of the big red flags are when you look at CVs; don't have them. Don't like formatting has to be clean, right? That's the very first thing that you see by opening the CV. Please be clean. If you trust your kind of design abilities, please go ahead and design a nice-looking CV, but if that's not your strong choice and if you can, if you put, I don't know, some example like 10 different colors on a CV and rainbows and whatnot, maybe we'll use a template online. It might be helpful. So I think it just doesn't have those red flags, and imagine just maybe showing your CV to friends, right, and asking them, I show it to you for 10-20 seconds; don't read through it; just scan it really quick and tell me your thoughts. Use those thoughts, yeah, maybe ask AI the same question. Of course AI doesn't have anything like I spent 10 seconds looking at it, and in 10 seconds, it scanned your CV completely and related it to the information on the internet, so it's insane, but still, you can even ask it, Imagine you're a human being. Ask what I would say just having a 10-second look at your CV. So yeah, to sum up, yes, it's unfair, but there are definitely pieces of advice I think it's pointless to go through. We can probably have a whole podcast on the CVs, what to do, and what not to do. I'm pretty sure you either did or are planning to anyways, and yeah, I think you should implement that. Don't be surprised that the process is unfair, and don't be discouraged by it. again we're all human I know that we try to be fair as much as possible, and we'll never be fair 100%. We'll probably miss out on a really good candidate in this process, I'm pretty sure.

TIM: That's a great perspective that you just outlined there because, God, how much of life's disappointments comes from a mismatch between expectations and reality, so if you go into it thinking that you will be treated 100 percent fairly all the time, you are doomed to be very sad, I would have thought.

ALEXEY: Very true, yes, yes, need to chill a bit, and so, yeah.

TIM: I feel like it's probably also worth differentiating between the things that you have some control over and the things you don't. You just outlined a whole set of practical steps any candidate could take to give themselves a slightly better chance of getting a foot in the door, but at the end of the day, as you say, if a hiring manager has 800 CVs and they're looking at them for 10 seconds each, by the time you've got to the four hundredth, you want to quit your job because it's so painful. Of course the 401st maybe isn't going to get the same level of attention; that is only human nature, and you have no control of that really.

ALEXEY: Yeah, I have an anecdote that I've read long ago online about it, and it's like a senior and a junior hiring manager talking to a junior hiring manager. The senior one comes to the junior hiring manager and they say, Okay, what am I going to do? I have this pile of CVs to look at. I can barely hold it, and a senior hiring manager says, No worries. He takes the top half of the resumes and he throws it in a trash bin, and the junior manager is like, What are you doing? We can't do that. What if, like, those people—we would like those people. What if they're a perfect fit? And a senior manager is saying, We won't do that. We won't like to hire unlucky people anyways.

TIM: I love that one. I feel like there's an almost identical anecdote but for university professors giving out their grades by throwing the papers across their stairs, and it's like ABCDE depending on which stair it lands on. That's a great one, the same kind of process.

ALEXEY: Oh yeah, and you have a curve right away. You don't even need to think about that for some, yeah, for all. exactly.

TIM: Yes, your approach is more moral and objective. As good as it can be, this is one area where I think AI could drastically help, and it is also slightly frustrating to see, personally, some of the legislation that spun up in places like New York and California, then also the fairly onerous GDPR laws in Europe, which I feel like some of them may be well placed and come from the right place but have perverse outcomes. So there's one in New York specifically called the Automated Employment Decision Law or something, which basically says you cannot use any kind of automated hiring tools unless you go through this ridiculously long list of steps.

ALEXEY: Yeah, isn't it exactly the opposite? Shouldn't you use automation to exactly prevent bias? And maybe you say exactly to prevent bias because humans, in their nature, are very biased.

TIM: Yeah, and I feel like the early if we keep doing like CV screening, surely it's a no-brainer to have a more guarded, more unbiased AI going through and scoring against a very specific criteria where it's not looking at someone's photo or their name or their blah blah blah; it's just giving you a score. Surely that's going to be better than a human reading the 800th CV that day.

ALEXEY: Yes, and I do agree that it might lead to a slightly better future in this regard, so here I'd also be careful to use it in a hundred percent of the cases. Sometimes, as a hiring manager, you would see small things in the CV that may be for AI and wouldn't be important based on your task definition. Maybe you forgot to mention certain things, and you would be like, I don't know, like a stupid example would be… Oh, I also started at this university, and I know this degree and those professors and those other people that put me on a very successful path. Should I give him a chance? Would you list all the universities that you think are good for AI to rank? Maybe not a tough one, but as a human being, that can be a bias, absolutely, but it's one of those I would say maybe good ones to follow, but then, of course, like you said, if you include photos and some personal information, then it might be that you can get rejected, and it's a hiring manager might not even recognize it right. Because if you look at 10 seconds for the CV, you might be completely on your automatic thinking process. Like I said, it's your 400 CV; you're not thinking about that. You just rejected it; you wouldn't even remember why you rejected it.

TIM: Yeah, and whether the thought is conscious or unconscious, if the outcome is still negative for a particular type of candidate, then I guess they wouldn't really care. There was a really interesting study in Australia a few years ago, and I'm sure there have been equivalent studies in Europe, England, and America where they would get tens of thousands of CVs and apply en masse to different roles with different groups of CVs. the only difference of which is that the names on the CDs were then tested for bias against ethnicities, so they did one in Australia where they had, yeah, they had group one, which was basically white first and surname; group two was white first name, Chinese last name; group three was Chinese first name, Chinese last name; and the first group got a 12 percent callback, and the third group got a 4 percent callback only. So if the only difference between you and another applicant is your Chinese name, you have only one third the chance for a callback, which is like

ALEXEY: insane. Actually, that's also something interesting for the future. I never really gave it too much thought, but now that you mention it, I remember like more than 10 years ago when I was applying to different roles, I was also reading online what are the best practices to beg them for CV crafting, and back then it was definitely recommended to include a photo. I was doing that; I was including my photo in the now it's completely the opposite now, and I've seen it from CVs, and I think it's a good practice people don't include photos, and I also recommend people not to include photos. Why do you want a bias to happen? So I'm wondering if something like that would happen to a name because, yeah, what you're saying is absolutely right. and I can't really see it happening in real life and companies, so yeah, maybe I don't know another hot take: how about as an industry we just decide names in the process? It doesn't matter until a certain stage, which would be interesting.

TIM: Yeah, that's A feature we incorporated in our product in pretty early days was a feature we called candidate cloaking, which was a way to look at people's scores completely anonymized, so no names, no email addresses, no phone numbers, no countries—like nothing—and in theory, at least, the way in practice, which is to say how we've used it, is Look at their score; if they've scored above x, this person gets an interview. So at minimum, a person could get a foot in the door. I could still be racist; I could still be sexist; I could still be completely biased. But at least we've given them a chance to get into the building. I feel like that's an improvement on the current situation.

ALEXEY: Oh, that sounds great. Yeah, it would be probably if somebody would try to implement the system in Carfax. I would be very careful because I would have this feeling like, Okay, what would happen? What are we missing out on? What if I can always put numbers on people, right? Can I always describe somebody well enough with an algorithm?

TIM: We shall see and Yeah, it's interesting as you say that the kind of recommendations Let's say a photo on a CV or not very thorough time also for candidates what they should think about, I think, is their market. So I've had the fortune to hire in numerous different markets. It is very common in the Philippines to put Roman Catholic as, like, very prominent, right at the top of the CV in India I've seen plenty of candidates put

ALEXEY: right

TIM: Their cast on their CV, their marital status would be really common. Whereas obviously in Australia you would never see that; I've never seen a photo in Australia ever, and so it is a bit sorry dependent upon where you are in the world.

ALEXEY: Yeah, absolutely, but then it again comes to research. You can research it yourself; you can ask language models. Yeah, I think it's a very good practice to do research if you want to relocate. You're applying for jobs in another country, in another culture; that's a considerate thing to do. Maybe research a little bit what this culture values and what you should maybe include or not include.

TIM: What about now thinking about soft skills and technical skills you mentioned almost just in passing, then that maybe we can't reduce a candidate to a single number, maybe there's some things that aren't as obviously measurable? I wonder if soft skills might be some of those and how you think about soft skills and technical skills and evaluating them.

ALEXEY: Oh yeah, that's another good one. I think that this one is particularly varied in terms of the level of the candidate. It's a slightly different approach for junior people to senior people, so for example, when you interviewed junior people, they might be fresh from university as they might have just a couple of years of experience. and I put a little bit more emphasis on technical skills because I want to get a feeling for what they already know, how far they got until this stage, and I trust them more to develop soft skills as they go because I know they just started their career; they might not know even what they should do or not do or how to approach different scenarios for senior people. On the other hand, it's almost completely the opposite. I trust that by the time you have 10 years of experience, you probably know your stuff. I can ask maybe architectural questions, but I would value much more even how they solve particular problems and what values they delivered for the business, much less if they know a particular library, if they worked in a particular framework, or a particular language, so senior people can learn a completely new language like in no time. But soft skills, on the other hand, they developed; they've had experience, and I think they usually accomplished it as a personality, and yes, it can be a debate: can a person really change or not? It would be more difficult for them to change skills, so there I would put more emphasis on really trying to understand from the interview how's the style of communicating or not. Do you like them as a person? Do you think they fit in the culture of your company and in the team culture and the team dynamic? Think about all that much more for senior people, as engineers and juniors can be more like you can, like changing them from a soft skills perspective. easier

TIM: That's a great distinction you've made then between the leveling and the expectation over whether or not the soft skills can change because typically I would hear people saying the soft skills are not fixed; it's harder to change technical skills; they are easier to learn, but the fact that the grads or early-stage career people maybe are more malleable and just don't know what they don't know They might not know the nuance of business communication and that sending a hundred-page email to the CEO maybe isn't the

ALEXEY: Yes. Yes, that's exactly right. Yes, so it gives them some slack for them in that regard, but then if you apply for a data role and you had, I don't know, a stupid example, like, Don't need types of joints, something like that, and SQL never touched Python, okay, what are you doing? Yeah, you can do it on a site; you don't even need to study it.

TIM: What about then thinking about measuring these different things because you mentioned in passing then in an interview that likability might be a factor? that you rank the candidate on are you explicitly having this as part of a scorecard where you're saying likability is worth 10% Let's grade them out of 10 on likability, or is it more

ALEXEY: No, I think for me it's more intuitive, and when I ask, sometimes I actually present in between on very different stages of the interview process. I sometimes choose to be directly on the screening and have a first call with a candidate depending on the role and depending on the situation if you want to distract a team or not. all that, or sometimes it was at the very last stage of the interview, so when the team is happy, HR is happy, everyone is happy, and we just want to get to know the person. That depends a lot on the stage of the interview process, I would say, but I think that's very important. That's what I usually ask the team when Okay, let's say I'm actually at the last stage, and they already had a call with a candidate. I like to know from them, from actually each individual that was on a call, if they think a person can do a job. That's perhaps the most important thing, and that can be scored and objectively verified. then two, if they think they can learn from a person, varies a bit on the level; junior and senior might have different answers, but I think if you bring somebody into the team and into the company, the whole team is convinced that they don't bring anything we can't learn from them, even if they technically can do the job. We don't need wheels in a mechanism that just spins; we just replace it once in a while. Again, we need people, and it would be what kind of development do you see in the team that you never bring people from, and the third one is do you like this person? Do you like them as a colleague? Do you like them? Would you like to have a beer with them or tea, coffee, or whatever? I think that's very important. If you can learn from a person and they can do the job, but you simply don't like them, you can't even explain it. Maybe they're nice individuals, but conversation doesn't flow. Interest doesn't flow; it doesn't match. Everything is always a tiny bit awkward for me. It's a flag to consider and just say maybe we're not a good fit, and that can be put to a number. It's more like a very subjective feeling. I usually would not rely on just my feeling when it's like that, but if I feel something and then a bunch of other people are saying, Also, yeah, you know what country to explain it, but I didn't really like them I don't know why that would be a no, so we want to have a definite yes when we make a hiring decision, and if it fits but we don't like them, then it's a no, so it's important to answer your question, and I don't think it's very quantifiable. It's subjective; it's biased, but at the same time, it's important.

TIM: I'm so torn personally on this one myself because, yeah, like anyone else, why wouldn't I want to hire people that I like? That's going to make working with them more enjoyable, but then also I feel like the lack of liking them could just be some kind of, if I really dug into it, would I just discover some way that I'm discriminating against them, or there's some subset of the population who are stuck? like I mentioned, if you were very on the end of the autism spectrum and you were very difficult to form relationships in the early stages, maybe you couldn't make good eye contact, maybe you were a little bit awkward, you hadn't had amazing socialization; I could imagine how, especially in an interview process where you're a little bit nervous, you come across as a bit unlikable maybe you put your foot in your mouth a little bit, say the wrong thing, so I just slightly torn what do you think about that

ALEXEY: excellent thing, and that happens to us on multiple occasions, and I actually do have a recommendation on what I think works in those cases: have another call, have another meeting, reflect a little bit on the situation, and even, like, to set if you start asking your question, was it completely fair? Was I maybe a little bit wrong in thinking that was the candidate too nervous? Those are fair questions, and you can test for those. Have another call. and I think you can even be honest with candidates when it comes to that; they would appreciate it. I would have appreciated it for sure, and you can say Hey, can we have another call? If you are at the same location, try to have lunch together. Why not see if a more informal kind of setting would help? It might, and then your opinion might change, so I think doubting is fine; again, we're not perfect. It's completely fine that after one hour, one hour and a half, you didn't get a complete picture. See if the candidate is open to having another conversation. You can say, You know what? Let's have a more informal one. see what they like, what they don't like, movies, whatever

TIM: Yeah, that is really helpful practical suggestions on how to deal with that kind of scenario. I wonder if this is also almost this debate or spectrum from A purely intuitive gut-feel way of hiring where I imagine it's like on one end It's like I met someone for coffee and I made a yes or no decision. To the other end of the spectrum, where you've measured everything soft skills technical skills scorecard This candidate scored 89, whereas this one scored 65, like everything just comes down to a number. I wonder if maybe I've dismissed a little bit the gut feel for being less intelligent than what it is, like if it's this thing that we've developed over hundreds of thousands of years, and we just don't understand how it works properly. that doesn't mean it's not telling us something It could still be telling a signal, not noise, is what I'm trying to say.

ALEXEY: Oh yeah, I do believe that. I think you have to listen to intuition and have to listen to your feelings; you don't need to rely on them a hundred percent. I think it's stupid you need to look at numbers and try to find objective methods, but yeah, the truth is probably somewhere in the middle, and I can assume different companies would choose to dial it. up or down according to their needs I can imagine if you have a huge corporation and you have hundreds of thousands of candidates for each role, maybe you cannot afford a Tinder style of screening when a hiring manager with the team just sits in every day looking for thousands of CVS. Maybe you want a system to first recommend what to look at, and that makes complete business sense to me. Maybe if you are a startup and you are now hiring your first team, you want to establish a certain company culture because you believe in certain values, and you want to find people that believe in similar values. You can't put it on a scorecard, and then your dial needs to be you need to have some objective measures where you'll probably value This personal feeling, such intuition, much more, and I also seen it like with smaller companies, at least in Germany, it's a very common approach where you almost always have a round with a CEO of the company, a managing director, somebody, and I think that's exactly what they do. Would they ask you types of SQL joints? No, they seemingly ask very weird questions, but all of them drill down to do I think a person fits culturally? Do I like them? Do I like them to grow with my company or not? Do they share my values with this kind of stuff? And that comes mostly from the intuition.

TIM: Yeah, and in those interview processes, I think it would be great if companies could maybe be a little bit more proactive in their feedback. I guess this is for any stage of the hiring process, but especially these a little bit more subjective stages, especially for the candidates who fall over. I can remember personally one example; this is one of many, but early in my career, going for a role with a financial modeling business and getting to the partner level interview partners that ours is great, like you're exactly what we need. Just come in tomorrow and have a chat with my business partner, though. like we've got the job, but just come and have a chat with him. I did a four-hour round trip and had a chat with this guy. Okay, I was in his office for 10 minutes max. All we spoke about was Tottenham Hotspur and his ACL injury and nothing else. Maybe I dissed Tottenham. I made some negative comment in favor of Arsenal or something; I can't remember. but something I said in those 10 minutes made me clutch defeat from the jaws of victory and lose that job, and I still have no idea to this day, so it was something intuitive, something I said, a certain value I demonstrated or didn't demonstrate, presumably And so I feel like if companies could at least provide a little bit of feedback, that would be helpful to candidates.

ALEXEY: I completely agree, but also I had in the past similar topics discussions with HR on the topic of how do you reject a candidate, and this is where I think we have again our laws have very good intentions protecting candidates, but it often backfires in exactly those situations, so you cannot reject someone based on I didn't like you; my intuition was not exactly fit. Yeah, I think you're fit for the role; you have the skills; everything is fine. I didn't like that you can't say that legally people can literally sue you, and that's when it gets a bit again. not exactly sure how it is in Australia, but at least in Germany that's dangerous obviously, like, it's among other things, like you cannot reject somebody based on their age, based on their race, whatever; it's a big no, and in all honesty, sometimes you do reject people based on their age, that maybe you are a small company, a small startup, and the whole team is young. and you are thinking about a team dynamic. You are thinking that you know what, but they have to work together for beers and pizzas, and they draw on the board and stuff, and this candidate has family children, great children, and that's not where the priorities lie. They wouldn't do that, so that will destroy team dynamic. Very I think it would be so fair to just tell a person, Say, you're a great guy, and I think you could do the job. We can't do it with the current team; it just won't work long term, and legally, absolutely big no. I can say that, so it's a bit stupid from my perspective. I wish we lived in a more honest world. but yeah we don't

TIM: Yeah, until then we'll have to deal with what we've got. What about if you think back to your career in hiring? Is there ever a time that you had to make a particularly controversial decision, for example, hiring someone who everyone tells you, Don't hire them, or some other point of tension in the hiring process that you could discuss?

ALEXEY: Interestingly, I think I never really, when it is when it's controversial, I try to get as much information to not make it controversial; that's my top priority, so I don't think I ever made a call where it was controversial from everybody, like people have doubts; it can be good; it can be bad. these are pros these are cons When there are cons, it's already, Yeah, you have cons; maybe you should wait to consider somebody that won't have so many cons. So I think, like, I personally am very careful when it comes to that. Okay, it somehow needs to be something like exceptional. I don't know if a person needs to maybe bring that, but that's hypothetical. I never experienced that need to bring such unique knowledge perspective to something so outstanding that you can close your eyes on whatever aspect is controversial, but then again, in whatever creative way I can think of gathering more information, more knowledge to sort out this controversy would be, I think, my priority and not really to gamble. I think you in business and hiring, when you deal with people, maybe you should minimize gambling. Don't play those games; this I don't.

TIM: And then do I gather that you're focused in hiring on reducing the false positives, so like making sure that if you do table enough with someone, you are like not 100 percent sure it's impossible, but you're very confident they're the right person?

ALEXEY: I think I need to be and the team needs to be excited about the candidate. They need to think things are great; we need to like them; they need to fit. Everything has to be perfect. Not everything, of course, is perfect. Some things you let go, close your eyes, because you don't care; maybe you value other things more, but you need to be excited. You need to have a really good feeling about it, and to me at least, making a controversial hiring decision would mean it can go good or go bad. I put my foot on it. I think again my gut feeling is telling me that's the right choice, and you do it, so I wouldn't. I wouldn't do it. I somehow would continue trying to get information, and if it's impossible, maybe we'll have four or five rounds with a candidate. What more can you ask for? You gather all the information, and you're still not sure if you are not sure. If you're not sure, it's a no. I think that's, for me, the case. I don't like gambling in the sense

TIM: And has that been a philosophy that has been driven by any times where you maybe earlier in your career, when you were an earlier hiring manager, you maybe weren't as accurate with your process and you hired someone who you regretted, and you're like, Oh, I'm not going to make that mistake again? I'm like, is it driven by personal experience?

ALEXEY: Proof yes and no. In my career, I can recall one case when we hired someone, and I had a good feeling about them. Oh, to be fair, the team also had a good feeling about them, and it didn't work out from the probation period, but we try to be fair in the sense that we have frequent checkups during the probation period, and you see how the person is going. You tell them very openly, Look, I think you are doing this thing well, and this is where you can improve, and this feedback gets As a probation period progresses, and in Germany it's crazy long—it's six months—so you can't have a fairly good evaluation of a person, so in the beginning it's like suggestions. and then when it comes closer and closer to four months, five months, it becomes critical, and then we usually tell a person also openly, Look, This thing is critical; please pay attention to that. Please correct it because those reasons, if you don't, I think we will not continue it, so we'll still see maybe for another month how it's going. If it's not happening, I'm sorry. and when you set expectations right, it works really well, and like I can recall exactly one case when it didn't for whatever reason, so we had to part ways, but yeah, I think that's also what probation periods are for. You, of course, try to be not to again, like I try not to make controversial decisions in hiring. If you go for it, then you need to be prepared to have a very close eye on the probation period and do those checkups and feedback sessions and again eliminate this controversy. Yeah, let's at least say that after the probation period passes, there should be absolutely no doubt. That's a great fit. If you have doubts during the probation period, that's a red flag.

TIM: I heard someone from Y Combinator describe it this way, especially for startups when you're hiring maybe your first few people: put a marker in your calendar after three months. If after three months the thought of this person leaving doesn't make you want to crawl up into a ball in your bed and not get out of bed for a weekend, then they aren't the right person because the bar has to be set so high, especially in those early stages. What I'm interested about is: Were you ever in maybe a hyper-scale-up company, like in 2020 or 2021, that had endless money? It was like, Alexey, you have to hire 40 people in the next three months, and under that pressure, did you ever change your trade-off and become a little bit

ALEXEY: Ooh.

TIM: risk averse

ALEXEY: probably yes, if that's the case. If you cannot afford to, if you need to grow fast, you have to take risks; that's what it's all about. Ideally, I wouldn't. I again would like to think of ways how to minimize this risk, but then, of course, that's the priority. We need to hire 40 people in the next couple of months. You'd have to make risky calls, and then you have to be transparent with your managers that you would have to correct it as you go, and some companies do, and that's exactly a very good question to always think about having the knowledge that I have today, like what you said: Would I hire this person again? They can be working for a couple of months, or they can be working for maybe a couple of years. If that's the case, there's no need to reflect on it, and maybe it's better for everyone to part ways, and if that's the case of these hyper scale-ups, you have to have those conversations with yourself frequently and make those tough calls frequently. That's probably what makes it very challenging in those situations.

TIM: And it sounds like it's almost a case of the hiring process doesn't stop when they sign the contract or on day one; it's almost like throughout that whole probation period, you need to be really checking in a structured way to make sure they're hitting your expectations because you still

ALEXEY: Oh, yeah. I believe so. Yeah.

TIM: It's over.

ALEXEY: Yes, yes, I believe that's the case because we're all human. We try to be objective during hiring, but we can make mistakes. Five of us can make the same mistake. We should be brave enough to correct it and fair enough. I think it's bad taste for you five people from the probation period, and they were thinking they were doing a great job, and they have absolutely no idea why it happened. But if you are failing it, you also minimize the emotional load on that because you are open and honest with the person, and they won't be questioning what did I do wrong.

TIM: Yeah, if you're firing someone and they are genuinely shocked, then something's gone awry, and I feel like maybe that's for first-time managers. Even in my own experience, one of the hardest things to do is to have hard, honest conversations that you don't want to have, especially if you're a conflict-avoidant kind of person. The last thing you're going to be doing is telling someone you know, "Oh, there's a lot of issues you've got at the moment; you have to fix them. That's an uncomfortable conversation for many people.

ALEXEY: It's very difficult; I know that for sure. I don't think I'm very good at it, and it's probably easier when a person is, if you imagine a person that is completely not fitting, a jerk to everyone, not doing their job. Yes, it's easy, but that's not real life; it doesn't happen quite often. person shows up is doing their job how polite good people to go around with, and then, like, it makes it so much more difficult because you need to give maybe a bit of feedback to, yeah, again, somebody you like from the interview process, somebody that you have a good feeling about, but now that you work with them as a science, more objective metrics, and yeah, very, I like, again, I'm lucky enough not to have that happen often, really lucky. But I can imagine, like in hypothetical situations, bigger teams, bigger companies, and tougher deadlines on all aspects—oof, yeah, that's a tough one.

TIM: Yeah, you say you've been lucky, but I'm sure it's also having a thorough process that you've outlined and doing things the right way and following the steps and measuring what you could that has led to the outcomes being consistently strong. maybe a little bit of luck in there but some good forethought as well

ALEXEY: Yeah, thank you. I hope so too, but yeah, luck is definitely there as well. I guarantee it. I feel I'm a lucky person in general, so yeah. I'm at the bottom of this pile from my anecdote before just because I'm generally lucky.

TIM: Alexi Thanks so much for your time and your insights. It's been a really great chat.

ALEXEY: Yeah, thank you, Tim, so much. Thank you for having me, and I had great fun. Thank you so much for the questions.