This AI Conducts 100 Interviews So You Don't Have To

Sean Weisbrot: Vol Goloshuk is the CEO of AI Champ, sales talent as a service. It's an AI that helps you to interview people and move them through the recruiting process. This is a very interesting conversation I had with him. We talked about what AI is capable of doing. As well as how AI can help humans because we believe humans still need to be part of the loop for now, at least, as well as the future of what this looks like and much more. So, if you are looking for a job or you're looking to recruit salespeople, you want to see this interview. Why is the future of business for you being involved in AI based recruiting?

Vol Goloshuk: The way I see the AI as evolving is that, um. It, it's a lot better,  than humans in analyzing data. And,  recruiting is,  pretty much about,  analyzing each applicant. So typically, the way recruiting is happening,  you have,  a, a,  recruiters,  with a scorecard. So, they analyze,  let's, uh.  for example, take like sales, sales roles. So, they look into things like experience, what kind of industries you're selling to. Is it SMB? Is it mid-market? Is it enterprise? there are a lot of data points,  which essentially are,  being conducted manually. So, each recruiter has,  typically like,    like a spreadsheet, and then they. They fill in the details and, okay, this person is,  you can score that that much. So, depending on,  conversational skills, experience and,  other things that they're being measured. And, it, it turns out that AI is pretty good at making,  analyzing data and making the decision. So,  if you feed this data to AI model, so let's say, uh. You've got 10, 20, or 30 or more applicants. Everyone has different skills, experience,  and   you need to qualify somehow. How do you select the best one? And,  and that typically has been done manually. And now AI is changing. The game is essentially allowing you to run at this data through a single, uh. Script and then give you immediate answer based on the parameters that you value within your company, whether this is a combination of cultural fit, whether it's,  experience, connections, or a little bit of both. You can set up a priority of each, as quality of each applicant as well. And typically, only the most senior recruiters were able to get this kind of information. But now with ai, it's just making things a lot easier, cheaper, faster, just changing the game.

Sean Weisbrot: Isn't it possible for the bias of the creator to prevent the really qualified people from taking the next step in an interview and rather be filtered out be by the AI based on that bias?

Vol Goloshuk: Initially when I was starting AI Champ, I was reviewing the existing. Solutions. And,  there was a lot of negative,   like conversations online from different applicants. They were like, oh, I were able to, I got disqualified. And then some others were able to,  to complete the process, but they just discovered that the AI is just,  it's, it's something is, is broke. And typically,  that's what happens with like early versions of ai. It's like. You can compare them with a junior recruiter,  somebody you just hired,  of a different role. They're hoping on this and they're like asking questions. They're not relevant or they're not prepared. You are asking the,  like,  asking them something,   that is totally not related to the role or experience. And then it could,  for somebody with that ex a, a great experience in the specific field, it, it's, it can be off-putting like you, you're like, oh. It's what are, what, why are we going in that direction? It's not relevant, et cetera. So, it's very similar,  with ai. So early, early days of AI recruitment, well, exactly like that, but it's changing.  and, uh.  right now, with better models, we were able to increase the quality of the experience,  and then essentially,  minimize those errors with getting disqualified by just,   some, uh. AI hallucination as we call them,  something in accidental.  but there's another question,  emerges from this,  as essentially,  how in some cases when the applicant might be amazing, but it's not matching some of the existing parameters. So, so in this way, how do you create those exceptions? When,   some talent comes in and, and then often,  what happens in the recruiting space when,  there is a great applicant, but AI is just disqualifying because of like some parameters that,  humans thought that it's, it's,  super relevant and we shouldn't consider anyone else but. With,  with some,   depending on the existing applicants. And we kind of evolved our process and then we adjust the, the role and then requirements just to have somebody benefit. And that's historically been like,  some human needed there to, to make this,   judgment call and analysis with existing applicants. So once this part is also connected with ai,  then yeah, we're gonna have a much better experience and, uh.  and better, uh  easier, smoother applications, better, quicker decisions, better feedback from, I

Sean Weisbrot: agree. It's important to have a human in the loop on something like this. I was involved in HR with a few different companies that I worked for as well as companies that I've owned. And to date, I've never personally used an AI in a hiring process. cause when I was involved in those aspects of running a business, it was before AI was available as it is now.  and I'm, I'm not. Currently in a position to be recruiting people. Therefore, I'm not actively using any tools, but from my background in psychology, I know that a human is necessary for recruiting because I. When an AI is solely responsible for taking in all of the applications, scoring them based on perceived,   requirements or metrics or cultural fit, when you don't have that human input, you can't really manage those exceptions. Like I may have hired someone in the past that, their hard,  skills test wasn't the best. But it wasn't as good as other people, but their personality was a better fit. And so, I would make that exception and hire them because you can't train for personality, but you can train for skill. So, if they had a good personality and they'd be a good cultural fit, but there, their skill level may not be so good, you just hire them with a lower salary and you spend that time to train them. And then you've got someone with a better personality and the same skills, even if it takes a few months to get them there. So, there's, there's a lot of these nuances that, that's just,  software development as a specific example because that's,  the bulk of my experience, but. I think when you don't have a human there, the AI doesn't have the emotional capability to understand that the company would benefit from having someone that may not have the same level of skill as another applicant, but may have a better personality fit. Just specific examples for you,

Vol Goloshuk: the human in the loop here is extremely important. And in making decisions or having still final interviews, and this is how we adopted this as well with, with the product that I built, AI Champ with my team. So, the final decision still, I held by a person. And then still the decision making is connected with,  with ai. So,  the difference is now we are making it a lot easier for a decision maker to, to make these exceptions. So,  once you have an experience,   like an interview with ai,  essentially every applicant can get a scorecard and then they can immediately see. What's going on? And then how, how did the whole interview happen? And then what is this person being scored against and what are their points? And then perhaps this applicant can see any immediate feedback and then see, oh, okay, I'm, it doesn't look like I'm fit. cause they require the specific experience that I just absolutely have. So, there's no even point having. Connections with somebody.  but in, in some cases,  these things are still stored within,    the system,  where,  the recruiter can, or the decision makers who's trying to hire somebody can still review these and then see, okay, so. Let's have a look at somebody who have,  better,  connection with like a better cultural fit, but not necessarily amazing experience so they can still see the applicants, their interview, and then they can,  see their scores and the,  the listen, the, the, the whole experience with the interaction with ai and then make an exception. So it's still possible. So I think the future with, with AI right now is, is still connecting.  decision makers, recruiters who are, or who are essentially making a decision in hiring, but also making a decision,  making it as easy as possible for them. So,  ensuring that they can see those exceptions and they don't need to do like,  20, 30 interviews or sometimes even a hundred interviews. They can have AI to do it for them. Then they can quickly see within like 5, 10, 20 minutes,  quickly go through the each application, see the,  people with,  more experience, but less cultural fitness and see whether they, they can be connected within their organization or vice versa. Perhaps somebody who's just starting their career, but,  not necessarily have those experience but amazing feedback, let's say. Based in the same city or,  have the same,  I don’t know, cultural background or some interest in like sports or a specific technology that the, the company is, is moving to,  to  towards. So,  so yeah, we,  I think the success in,  with building AI products lies in the connecting these things together and letting people make better decisions,  with  hiring and. Having those options brought to them,  in the easiest form possible.

Sean Weisbrot: When I used to conduct interviews, I typically, and this is for hiring, not for podcasts, but with podcasts, I typically have no script, and I, I go as natural as I can, but when hiring, at least in the first interview, I typically have a set of questions that I feel I need to know the answers to from the person, so they're standard. And then sometimes I'll go off script, depending on if they answer in a certain way and I'll go,  I, I need to know more about that. Right? So I'll, I'll, I'll look at each situation. How can an AI be trusted to stick to a script, but then also be capable of going off script in order to know more about that candidate? In a way that adds value for the recruiter to be able to better assess that candidate.

Vol Goloshuk: What we have built is essentially,   we can get AI to get specific answer. So, so let's say, um. And,  the, the questions with like, experience, some people or candidates may not be able to qualify their specific experience themselves. And then this is what the recruiter essentially is for. So essentially asking specific questions. So,  let's look into an example with,  like, uh. Hiring a sales executive with,  like for the product in like on customer onboarding,    like a space.  so then you can ask,  or configure,  like an interview to get the specific answers. And then if you're not getting these answers, you can ask the questions that are connected to that. So if the person has mentioned something about onboarding. Then you have,   the, the interviews is related to that. So then AI will automatically is like, oh, tell me more about this experience. Oh, this could be relevant because of the companies is in this space. So it can basically,  ask specific questions. So, so, so you can program now the interview process with high level things. It's like you, you need to figure out,  like their experience,  their, um.   achievements in the universities and then the AI can basically ask these questions. So let's say,  in the universities, some,  sometimes the score of like specific,  like subjects may not be enough, and then you can ask like whether they have. Whether they had participated in any like competitions or,   any like Olympiads or anything else that's been organized, but in different countries, cities, or states, it's just there are so many other, like different unique things. Then you, you, the AI can basically qualify those and then ask more,  unique questions to each case, depending on the.  the, the applicant's experience. So now it's actually possible. And then with AI you can just define the high level,  task with like, this is what you need to figure out. And AI will the interviewer from,  in,  yeah, in our case,  will essentially,  try to dig into the specific,  part of the candidate's experience to be able to get the answer. So yeah, now it's, it's possible.

Sean Weisbrot: You had mentioned earlier that the applicant is able to get kind of access to their scorecard after the interview to see if there's a chance of them getting the   to the next step. I think that's really valuable because companies usually never tell you why they're not gonna hire you. Just like women never tell you why they don't wanna continue dating you. Um. Which is a personal experience and  it's like I wanna get better, right? If there's something wrong with my personality, I need to know, or if there's something wrong with my experience or how I express myself, I need to know so that I can get better at this. And the only way I'm gonna know is if people are willing to give me feedback. And so I, I like that you're doing that. Another aspect of recruiting and I, I used to date a woman who worked for a company and she was personally responsible for hiring 50 to 80 people a year for the company. So I got to see through her eyes the different conversations she had ongoing with the different applicants. And this was a few years ago, so before AI was available to be used. And she was actively managing all of these conversations and their negotiations. cause these were executive level positions she was hiring for. So,  some of the people had to deal with relocation packages and benefits packages and things like that. I. Have you considered having features that analyze market salaries based on the position to give the applicants feedback on what kind of salary or package they should be like negotiating for? So like you, you have,  a product for the company to recruit, but you also have a product for the applicants to get better feedback and insight so that they, they can get the best deal for themselves when they do have a good fit for a job. Hey, just gimme 10 seconds of your time. I really appreciate you listening to the episode so far, and I hope you're loving it. And if you are, I would love to ask you to subscribe to the channel because what we do is a lot of work, and every week we bring you a new guest and a new story, and what we do requires so much love. So that we can bring you something amazing and every week we're trying really hard to get better guests that have better stories and improve our ability to tell their stories. So your subscription lets the algorithm know that what we're doing is fantastic and no commitment. It's free to do. And if you don't like what we're doing later on, you can always unsubscribe. And either way, we would love a, like if you don't feel like subscribing at this time. Thank you very much and we'll take you back.

Vol Goloshuk: It's like a marketplace.  from one side, applicants wanna get paid,   in,  yeah, want to get rewarded in a maximum way for their work.  which is,  how, how it should be. And from the other side, you have business who is running,   a company and want to get maxim  value,  from  every, every team member essentially to, to, to, to get the company to the next level. So,  and then in the middle, then we need,  everyone making a decision. So do  you have multiple companies and then,   applicants are looking Oh. Can I,   so,  like, to, to answer your question is, is, is, is not specifically by like, salaries expectations, but it's also could be more for,  what future can you see,  within this company and,  what potential cause At the same, at the, at the very start. Um. Like,   some less experienced careers then,  it's, it's, it's very difficult to,  to,  to com combine those. So whether the,  com,  like,  companies are essentially going to look into somebody with like more experience, but they also look, need to see what who else is available to with that experience. And then sometimes if you're away off the market, then yeah,  but I would say the easiest way is look. At the,   there are like websites to, to see like average,  expected salaries,  on, on, on the market. But I think for the remote roles it's difficult because we are just equalizing the world. And then sometimes,  if you're based in developing country, you can. Create value to,   a company in, in developed market. And then things,  changing. So it's not only where are you based, it's, it's actually what value can you bring for,  for the, for the salary or benefits that you're expecting. So, yeah. So there, there are a lot of,   yeah, there's a lot of points there where, uh. If I was an applicant, I would look into like a, a future with a company and then see if,   potential. And then with within the salary, I think it's, once you prove your value, it's  no brainer,   any, any company would want to,  keep you,  whatever the, the cost.  as long as you know the company's moving to the right direction, growing, and, uh. So it's, yeah, it's like never ending battle,  with a talent.   it's,   everyone wants to, like a company wants to get,  best deal from, from applicants and then applicants wanna get the best deal for themselves, for their future in career. So, and there's all, all these parameters that everyone is looking to, to match. And I think actually AI can. Can help each site,  to make a better decisions.  and that's where,  we can make a better experience for applicants. cause they can see, oh, okay, this is great, and then,  I need to improve here, or there is no chance I can get this,  because my score is just too low.  but  there's never answer yes or no because you never know you could actually get the job anyway. And then on the other side, also, the companies who need to make a decision quickly, and there are so many applicants. So, the whole decision-making process can be as quick as,  like hours or even minutes rather than previously you had people in the middle trying to score, analyze, and review and,  schedule reschedule at least,  interviews and all that. So, so yeah, I think AI is, is gonna help each, each side,  a lot with, with,  securing better opportunities.

Sean Weisbrot: in an era of deglobalization and inflation. Ai, why should companies still look to use an AI to help them to make decisions when they're also thinking about cutting costs and using AI in those positions that they might be, that you're trying to get them to hire humans for?

Vol Goloshuk: A lot of jobs will be automated. I mean, it, it's, it's a no brainer, I think,  and,  the way, um. We're seeing this as very similar to a case with accountants.   in the early days when computer got,    became popular,  there were,   like a, like software which allows to recalculate all of the, uh. Accounting,  balances and then in, in like automatically. And that was a pretty much the job of accountant before the computer was popular. So, and then all the accountants were like, oh my God, what, what I'm gonna do,  I used to do these three calculations on the different,   balance sheets. And now AI in spreadsheet, you can simply change number every, it's automated. Everybody else. That's it. I'm, I'm, I'm down. What do I do now? So, um. So, the same is happening here as well. I think we're just gonna focus on creative, better,  better,  valued work. So,  then the, the role of the.  like a basic knowledge base worker would, would be automated. And then we, we are just gonna have to move to the, to the more of creating value with like what do we do now with this? So same as with accountants. So now we used to recalculate these numbers for different spreadsheets. Now we can bring some more values-built software. So, the way. Now I see this as very similar with, with ai. So, all of the,  you used to need somebody to write the code. So similar to an accountant in nineties. Now in, let's say in three or four years, there won't be a need for somebody to write the code. You're just gonna be in line,  describing what you're looking for. So, I'm looking for this application. Is it gonna maybe more like interaction with. With,  with a person trying to figure out what, what's gonna be the most,   impactful as a  thing to create right now. So. So, yeah, I think it's,  yeah, it's,  in terms of hiring, it will also change to having more senior roles. And,  if we, if we look into sales specifically, it's not gonna change,   dramatically because I see this as people will still be,   buying from people. And it's the same with like,  playing chess.  computer plays better chess than humans, but. Do we care? I mean, we do, but we still have chess, chess,   players. And then, humans are still interested in playing chess with other humans. And it's,  it,  and I think it's still gonna be the same with,  with,  many roles that we have. It, we, we still gonna interact with people and,  we, knowing that AI can do some role or even things even better, but there's gonna be better experience with,  with the human connections in the future.

Sean Weisbrot: My bet is that in five years, my agent will be buying from your agent and you and I will never speak, but my money will change hands. And you, your agent will teach my agent what it needs to know.

Vol Goloshuk: I think I would agree with you here.  we all gonna have our agents, they're gonna interact with each other, gonna make decisions, and the transactional sales are gonna happen this way. And, what will happens then in the future?  in sales, for example,  this, the, this experience with people will move enterprise and,  if you need,   like,  interaction is, is more about who,  what,  what connections can, can you bring and,   and, and  other things other than just, uh.  making a decision. But,  but yeah, I think transactional sales is already happening. I mean, we, we don't need to have meetings for,  like $5,000,  a decision. But then,  things are gradually moving, so then it's just gonna move to a thousand dollars and more. And,  I think it's,  in the future, uh.  in the sales example, we we're gonna have the  talent,  in the more interesting deals with,  more stakeholders,  than decision makers that are happening, not just by one person. And then even at that  phase, we're still gonna have AI help assisting us. Way. See for example, in sales, every enterprise sales guy will have a sales engineer more, more likely. Some sort of agent, which hops on a call where there's no gonna be, there won't be any questions, is like, oh, can you, does your software does this and that? Oh yeah, let me check with my team and find out there's gonna be immediate instant.  AI was like, yeah, we can do that. We cannot do this. Or.  vice versa. So,  yeah, the, so the world is gonna be more interesting and in, in the enterprise space, but yeah, I agree with you. Like,  any mid,  enterprise, mid,   mid deal, mid-size,  deals, mi  very likely to, to, to get automated.

Sean Weisbrot: What's the hardest thing about running this business

Vol Goloshuk: to ensure that every experience is amazing? First from, uh.  client and also from applicants. Because historically this has not been amazing experience. This somebody was always unhappy either like no feedback or bad quality candidates, or not relevant candidates, et cetera. So, the hardest part for me is to ensure that every side now is happy.  they're like, okay. I, I got who I was looking for and on the other hand, or if I didn't, I, I am clear why and how do, what need, what do I need to do to, to, to change,  change to, to make a better decision. And on the other side as well, for the applicants,  getting clear feedback, knowing exactly what happening. Am I getting, getting hired here or should I just,  spend more time or not? That's the hardest part because that requires a lot of like technical,   calibration of the technology and,  a lot of prompting and testing.  and  yeah, I would say I'm gonna be focusing on this,  yeah, most, most of my, time,  for the,  next,  few quarters, that's for sure.