How To Become A Talent Engineer (Recruiting's New AI Role)

Speaker A: In the last 12 months, has AI actually helped you make more money? Has it actually helped you make more placements? Or is it just been a lot of noise? I know we just got done with the AI Recruiting Summit, which was absolutely amazing, but here's the thing. I'm partnering with Seb Sharp, who's going to walk you through how to build a bunch of tools. Two sessions, August 25 and August 26. You want to make sure that you get signed up for this. He is going to show you how to build an autonomous lead agent. He's going to show you how to build a placement agent. He's going to show you how to connect your entire tech stack, your ats, your CRM, all your sourcing tools together, how to build a free website to save hundreds of dollars a month with true infrastructure. And here's the thing, these are also going to have AI assets that you get to keep after these two days. Recordings, templates, and a week of Slack access to Seb. So to answer questions as you're building everything for yourself, walk through, build along, and then a week of access. You do not want to miss these sessions. This is the perfect follow up to the AI Recruiting Summit. This is your chance to, like, literally cut down a lot of your costs, uh, do all the things that you've been talking about and actually make money from AI. This is going to be awesome. Make sure to get registered. I will see you there because I'm going to be in class along with you while, ah, Seb's doing this teaching. Coming up on this episode of the Elite Recruiter Podcast.

Speaker B: Things don't happen to you. Things happen, and it's how you respond to those things that dictate what happens next. Build yourself an cp. Build yourself the ability to pull information when you need it to be able to get information on companies. Like, if I was on the agent side, I would want to arm myself with as much data as possible and then use that different ways depending on the client that I'm working with or what they're looking to do. Welcome to the Elite Recruiter podcast with your host, Benjamin Mena. Uh, where we focus on what it takes to win in the recruiting game.

Speaker A: We cover it all from sales, marketing,

Speaker B: mindset, money, leadership and placements.

Speaker A: Atlas, uh, is a crmx, a CRM that actually understands context. It captures everything you say, hear, read and write. Every interview, client call, LinkedIn, message, email and WhatsApp automatically. Not because you spent the time entering it. It's because Atlas is always listening. Then it goes one step further and tells you the next action to take. So when you need to fill a role, people search, ranks the best candidates and tells you why. No digging required. The same memory turns BD into a short list of who to chase and why. So you win more. Clients need to see your pipeline, speak to your dashboards. And Atlas builds a view automatically tracking whatever matters to you, whether you place permanent or contract. Atlas covers both with a new contract suite. That means you never have to chase a timesheet again and live margin visibility on every single placement. And with its new MCP and API, Atlas plugs into your, your LLMs, Claude, ChatGPT and the rest of your stack. The low value work that keeps you from billing, Atlas does it for you. Atlas customers right now are saying that they're seeing a 50% higher candidate response, a 35% increase in new clients, want 15 hours saved every week, monthly billings jumping 85% and agencies hitting 150% of their annual revenue target by building their business with Atlas. If you want more information, check out recruitment with Atlas. I'm excited about this episode of the Elite Recruiter podcast because here's the thing, like this new position I think would have been actually a dream job for me years ago. And it's, it's something that I see as a major impact both for in house recruiters and agencies. And we're seeing this at a lot of, um, agencies, but a lot of this is actually being outsourced to another organization. I think this kind of role is going to be brought in house to just every, almost every single boutique firm soon. But here's the thing. It's a brand new role. Everything's now being designed. The experts are the ones actually building it right now. And I'm excited to have actually one of those experts, one of the very first people with us, part of the A16 fellowship for the talent engineering program. So, Dan, excited to have you here, brother.

Speaker B: Uh, I'm excited to be here too. Thanks for having me on the show.

Speaker A: Let's just dive in. So you, you built AI for talent teams for a living and you told me that nobody's like actually quite proven. It's moving the needle. How are those two things the same right now? Yeah.

Speaker B: Starting with an easy one. I see. What I will say is that I think that, I think we're in the early stages of all of this. And I know it may feel like we've been building for months and if not years at this point, but, uh, I think when we look back on this time in five or 10 years, we're going to realize just how early on in this journey with AI we are not even just talent engineering or recruiting engineering, but I mean just AI in general. And so I think we're still figuring out what, what progress means and uh, what ROIs we're looking for. Does that mean specific metrics are getting better? Does that mean candidate experience? Is it more subjective in terms of the progress that we're making? And I don't know, I think that there are some teams that are certainly seeing progress in one, two, perhaps all of those, and maybe even some that I'm not thinking of or mentioning. But I think right now what we're seeing is still a lot of experimentation and we're feeling through the dark to see what those things are that will actually move the needle. And not just a little bit here, a little bit there. But truly, what paradigm shift are we looking at that will allow us to do this job in a different way than we've seen it done in the past one to 30 years?

Speaker A: And one of the things that I'm seeing on LinkedIn, and actually if you want a good laugh, like 30 minutes before we sat down, I was in quad working on something, I was in lovable working on something. And Everybody's hopping on LinkedIn. And I think this is one of the reasons why everybody feels so far behind is you're seeing like screenshots of uh, like, hey, I just built this yesterday, I just built that. Is there something wrong with that or is this causing more issues?

Speaker B: I think inherently there's nothing wrong with that. I think that the way most people are doing it right now is not very beneficial to the general public. I think what we're seeing right now is we're getting a lot of folks that are building and they're on the computer. Maybe they're at home, maybe it's their work, whatever they're building things, they're creating as they go. They're again feeling their way through literally in the moment. And they build something that works for them on their local host, on their computer, and they decide to screenshot it and then they put it on LinkedIn and they talk about how they built this brand new tool that's going to change the way their company hires talent. And it's maybe they built their own ATS and they can get rid of Ashby and they get rid of Greenhouse and then they're going to get rid of Juice Box next. And I don't think anybody's truly doing that. I think that what we're doing is something I'm, um, unfortunately referred to as ego building in that we're trying to show everybody that we're as far along as everyone else, or at least as far as we think they are. And so we're building a thing and then sharing it. And maybe it makes you feel better, but I think the problem with that is that one. Is it actually going to work? Probably not. And even if you manage to host it somewhere and get it past all the security reviews and all the teams sign off on it, what's the adoption like? And even if you get adoption, what needle has it moved? So, uh, and so if in six months you can look back at the data and say, yeah, I built this thing, we went through this, this and this, we got sign offs, we talked to stakeholders, we got buy in, we built it, it's hosted it, and now we've seen a, uh, 40% drop in time to fill. Awesome. Share that. Tell me, tell everybody. I want to do that too. But we're not doing that. We're doing this thing where we just share a thing we built yesterday that's probably not going to be used in three more days. And it's making folks that aren't doing that feel like they're even further behind. Like, uh, oh, they just built an entire sourcing engine and they use 17 terms in their posts that I've never even heard of before. I'm so far behind. And it's like, no, they just did a thing and screenshotted it. Like you could do the same thing if you wanted to and have Claude write you a post that says 17 things that no one else understands either. So I think all it's really doing is it's making people feel more behind instead of sharing the struggle or sharing what you hope will go or even just building it, but then being like, hey, I built this folks. What do we think? Like, do we think this is going to make a difference? Do we think this is helpful? If anybody has questions, let me know. And that's how I try to approach things is like I don't know everything by any means and I will always ask questions. I hope that things I build will work, but I don't ever want to make anyone feel like they're not as far, uh, ahead because I'm doing this thing and they're not doing that thing. So I think we're just being a little, a little toxic for ego's sake right now and I think we'll get through it, but it's just a little Unfortunate for folks that are not at that stage yet.

Speaker A: Let's take a step back. What the hell is a talent engineer or recruiting engineer?

Speaker B: Oh, uh, man, I was hoping you were going to tell me on this show. I'm still trying to figure it out myself. I think there's a lot of definitions that people kind of throwing around. I think there's a lot of definitions that could be accurate. The way that I see it, I don't have a firmly ironed out Wikipedia one line description of what this is, but I just think of it as somebody that has both the TA lens and that has a deeper understanding of the technical side that can approach problems for talent in a way that people in talent have not been able to approach those problems with before approaching it through the lens of engineering or through the lens of building. I don't like to say that we're solving people problems by using engineering because I don't think you can solve people problems just with more tools. Uh, I think people problems will always have to be solved by people and engineering problems have to be solved by engineers. I do think there's a little bit of overlap in that Venn diagram. I think that one can be used for the other and I don't think you can't have overlap. But I think it's just, it's folks that truly understand, come from one side or the other, but have done a lot of work in the other field and they get to approach things and rethink things and build tools for recruiters or for themselves. They will genuinely not to use this term, I don't use that much. But apparently today, um, shift the paradigm, like change the way we are doing things permanently. So yeah, there's the very quick, you know, 122nd answer.

Speaker A: I think it was like about two years ago before the first like AI summit. I was like, hey, I have this idea of like a recruiting engineer. Man, I was so wrong two years ago on exactly what this rule was going to end up being. I thought the person was going to have to like do all these things and I think Claude just broke down some barriers and made it easier for people. So much easier. But anyways, that uh, we'll talk about that later. But real quick, 30 seconds, self introduction, Dan, so that people know exactly who they are listening to.

Speaker B: Sure. So yeah. So the 32nd life story from Toronto in Canada. Born and raised there. I was a jazz musician from the. Well, I was a musician for the age of three. My parents put me in music. I never wanted to do anything else. In my life other than music. I went to, to an arts high school. I went to college for music, played jazz, lived in Toronto and played jazz, moved to New York to play jazz. And then many things happened. I stopped playing jazz. I went through many different careers, uh, pivoted into I think what a lot of people do, sort of hospitality. I bartended in New York for many years. That led me to work, uh, in wine. Specifically I worked both, uh, managing a wine bar as well as working at a wine retail store. Uh, I got certified in wine through WSET and taught wine classes and wine tasting to people. That led me to meet a really good friend of mine, still to this day, whose name is Aaron. We opened a CrossFit gym together. You know the natural path of jazz musician to wine, to CrossFit like everyone does. And so we opened up a gym in Brooklyn, ran that for many years. Crow Hill CrossFit. Probably to this day the most fulfilling career I've ever had and probably ever will be. Just to truly see like people's lives changed. And then, uh, that kind of dovetailed. After I had a kiddo, my wife and I were trying to figure out like what would be long term sustainable because it's certainly not running a brick and mortar business in New York City. And the obvious solution was tech recruiting. I mean again like it's just this the natural path, right. My wife work had worked at Etsy for many years. She was uh, an HR BP there. So she's been in tech for a long time. And so one of her friends just helped me understand that my resume probably does have a lot of like transferable skills, which I didn't even know what that meant at the time. And uh, Shopify agreed. They hired me when the job on my Resume previously was CrossFit gym owner, head coach. And they were like, you should be recruiting technical leadership at Shopify. And I was like, cool. I don't know what that means, but cool. Uh, and then yeah, spent some time at Shopify, was laid off since it's how to telecommunications, uh, tech company called Calix was laid off, uh, boomerang back to Shopify when they asked if I'd come back. Spent a couple years there and recently past five, six months, landed at Zapier in this awesome uh, role.

Speaker A: And you're also part of the fellowship program for the town engineers. Can you talk about that real quick?

Speaker B: Yeah, uh, so a 16Z, um, Andreessen, uh, Horowitz. They've done some interesting fellowships with things like forward deployed engineers or growth Engineers. And uh, I think design engineering is another one they've done. And they reached out to me a couple months ago because they were considering doing one for talent engineering and they just wanted to sort of chat with me about it. And I didn't know anything about kind of what these fellowships were, didn't know they were doing them, but had some really great conversations with some really great people over there and they decided to go forward with it. Um, so I gave them some recommendations on some folks they should talk to that I think are doing some really cool stuff in this space. And so, yeah, so they sort of officially launched it. I guess they will. They launch applications. And I guess they've. I don't know if they've told people who's in it yet. I don't know who. Like, I know some of the people that are founders, but I'm not sure some of the other people that are joining. But first dinners in a couple of weeks. Beyond excited to get to meet some of these people in real life and sort of chat through it, what they're doing and how they're seeing the space. I think everybody has very different ideas, but honored to be asked to be included in such a, amazing, uh, group of people. Like, I very much. I said to my wife, I was like looking down the list of all these companies and people and I was just like, babe, one of these things doesn't belong. That's me. And she's like, you belong? I was like, okay, maybe. But, uh, we'll see. So very honored and excited about what that's going to lead to.

Speaker A: Okay, let me take a step back. At what age did you start recruiting?

Speaker B: My non linear career path led me all over. I didn't even know what the tech industry was really. And then, Yeah, I was 40 before I even got into tech and was recruiting. I remember I heard the term org chart at work one day and I didn't know what that meant. And so I went to. I was like, hey, can you just tell me what an org chart is? Like, just draw it. And she's like, well, it's gonna be different. I was like, no, no, no, no. Like, just like an org chart. Like not specific, like who's in the roles, but just tell me what org charts tech companies use. And she's like, no, no, no, no, no, that's not how this works. Uh, and so that's the level of knowledge I went into like, or 40 with. I, I knew nothing.

Speaker A: You told me something that I think a lot of recruiters struggle with. Whether you're in the agency seat, executive recruiter, staffing, even internal TAs, dealing with imposter syndrome, because you actually had a pretty non linear path. You dealt with imposter syndrome, but now you were sitting there, you told M me that changed and that became your edge. What flipped?

Speaker B: The short answer is I don't know. I think, uh, obviously imposter syndrome is very prevalent. I think it's only going to become more and more prevalent as AI accelerates and people are going to feel left behind or they do one thing that ends up being cool and people start to pay attention to them and they're like, I just stumbled on that thing and now everyone's looking up to me like, I know what's going on. We're going to get a lot more of that happening. I think for me, the people that I worked with and similar to the A16Z fellowship, like I'm going into this room with people that have been in the recruiting and ops space and engineering space for 20 years from like Airbnb and OpenAI and anthropic and all these places. And like, it still certainly hits me there. But I think what I've realized is that my lack, like my lack of experience in this industry can be a strength. And I always want to walk that line very carefully because there will always be things that people will know that I won't know. That would be just historical context and knowledge and they. And I don't know that and I never will. And I think that I kept telling myself that I had to know that stuff that they knew to be good. But there's no way to live another life, to get up to speed on everything that they've experienced. Where I saw the strength was that I get to look at things from a perspective that they don't because they may have these ingrained thoughts and behaviors and patterns and I don't have those. So when I'm asked to do something, if I have a greenfield project, for example, I don't have 20 years of experience telling me what this thing is supposed to look like. I have no years of experience in operations. And so I'm like, uh, oh, I don't know. What is talent intelligence? When I was asked as a year to sort of build this function and I was like, okay. And then in the back of my head I'm like, what does that mean? What does build a talent intelligence? What is talent intelligence? But I got to think about this from a different perspective. I think that what I ended up building for this was different than what most people would because they wouldn't. They would have more of a, uh, sort of this trail has been walked perspective. And they know what that means, whereas I don't have that. And so I'm just going to do what I think makes the most sense. And hopefully it makes sense.

Speaker A: Most people know that. I absolutely love wine. I gotta ask you, being a sommelier, uh, you taught wine class, worked at a wine bar, I'm hoping. I actually went to that wine bar, actually. Which wine bar was it?

Speaker B: Uh, Brookvin in Park Slope. 7th Avenue between 11th and 12th Streets.

Speaker A: Okay. I think I actually have been there.

Speaker B: Nice. Still. It's still there. One of my closest friends, Brian. Brian. Uh, bar, favorite place.

Speaker A: We hit up almost every wine bar in New York City for a while. So every trip we were just like bouncing around. Anyways, a whole nother story. Question for you on wine, your favorite wine region.

Speaker B: The answer is going to be, it depends. It's like asking a musician, like, what's your favorite favorite artist or genre? I view wine very much the way that Duke Ellington viewed music, where his perspective is, there's only two kinds of music. There's good and bad. And I agree with that for music and I also with that for wine. I think that not to get too off topic, the wine regions that I tend to love, like the wines that I really think about that, like, stick with me, are the, are the weird ones. They're the interesting ones. So it could be, maybe it's Prayer Rat in like northeastern Spain. To this day, the most interesting one I've ever had is a 10 year aged rose from Rioja. Fascinating aged, uh, in oak, actually. So it's, I think it's like two years in oak and rose are typically meant to be consumed within five days. But like a year, you know, that's sort of the vintage. Like if you're going to drink rose now, probably last year's vintage and that's it. But drinking a 10 year old rose aged in oak, I mean, it was just weirdest wine I've ever had. But it, I still think about that to this day. So, uh, I love those regions. I mean, I love anything from like a good, A good Burgundy. Uh, yeah. I mean France is full of amazing stuff like Bordeaux. Left bank, right bank. If you don't like Merlot, I guarantee you like Merlot. If you drink the right Merlot, you know, and that's what. Oh yeah, that's what Bordeaux is known for. Um, so yeah, I could go down that road forever.

Speaker A: But yeah, well, Actually does kind of flip into my next question. It's like between the jazz gigs, the wine bars, the wine classes, the CrossFit gym, like what did any of that teach you that you use to something that you use actually in your recruiting world?

Speaker B: Yes, absolutely. Where to start? I mean, I think the, the through line is that everything that I've ever done is people first. Right? Like you can't be a good coach if you don't understand the person. You can't be a good bartender if you don't understand the people that you're serving. You can't be, you know, none of the things that, that I've been good at, uh, can you do if you don't understand people. And recruiting is like that. Even with all the AI and the building and the technology that's coming out, I still think that it's uh, even more so now. It's important that we have the personal touch. I think it's also taught me, I think one of the biggest lessons from owning a business is it's very easy to be dogmatic about things and to think that the way you do them is the right way to do them because of X and Y. And maybe there's valid data to back that up. Even because you've had success and you've grown your business or you've done X or Y. But it still doesn't make it the right way or the only way to do it. And that's a lesson that I've learned the hard way many times. But it's, I think it's finally sunk in now. And so, you know, it's. When I look back at, at sort of the dogmatic approach I used to have towards things like nutrition and you know, people doing fitness outside, you know, outside of CrossFit, like it's really easy to get focused on that and forget that people are people and everyone's going to be different and what works for you might not work for someone else. And you need to be flexible in your approach to all of these things. And I think one of the things that I really loved about Shopify, one of their values was strong opinions loosely held. And I never heard that before Shopify. And I know that's relatively common in tech, but I never heard that before. And I was like, this is what I love. I want to be around people that have opinions and thoughts about things and that want to tell me why they think they're right, but that also have the self awareness and the openness to want to hear my opinion and let's have a conversation. Like, because maybe the answer is a combination. Maybe you are right. Maybe I am. I don't know. Like, let's. Let's find the right answer. Um, and that's something I used to say a lot when I was recruiting Shopify was like, I want to be in a culture of people that want to find the right answer, but that don't care about being right. And I think the best companies, the best talent teams, are full of people that share that attitude. I'd rather be wrong, find the right answer than be right, and then have it be the wrong thing for the team. So all these lessons and probably a hundred more that I could probably put in a book at some point, but those are really two of the big ones.

Speaker A: So you got hit by the Shopify layoffs. And I know a lot of recruiters, both agency and internal, over the last few years have been nailed by a layoff. You posted about it publicly, like, while it was raw, what happened next?

Speaker B: Yeah, it was. I mean, I was. The irony is that I sold my gym in Brooklyn and left New York because I want. Because we had a kiddo and I wanted to find a more stable job. And so the irony that moving to tech was the opposite of that was not lost on me at the time. And, uh, still not lost on me to this day. The way. So it happened for a bunch of years leading up to that point, and still to this day. But I love reading. I'm a big book person, a big philosophy person, and I discovered stoicism a bunch of years before that. And just, uh. And whether or not people are into stoicism or any philosophy, it doesn't really matter. But I think one of the biggest lessons that I've ever learned is that, generally speaking, nothing that happens to you is good or bad. It's your opinion on the thing. If you have an opinion on it. It could be good, it could be bad. But things don't happen to you. Things happen, and it's how you respond to those things that dictate what happens next. And so you could be complaining and you could go on LinkedIn and bitch about your previous company or whatever. And I just. I don't think any of that's productive. Like, I want to control the things that I can control, and I can't control getting laid off, but I can control how I react to that. And so I took the day to be sad and to talk to my friends in the Slack channel and to, like, what's up with my, like, super close Group of five, uh, or six recruiters that we all onboarded together and then we all got laid off together. And then the very next day I woke up, I was like okay, cool. How can I make this into the best thing that's ever happened? And this sounds so cheesy but if you ask my wife, literally this is what I said to her the next day. I was like this because we had this move to California, we were living in Toronto at the time and this moved to California plan. And I was like, I just got like whatever, however much severance I got. It was a very generous, I don't remember now but it was a very generous severance back. I was like if I can land a job in the next four weeks, we've just paid for our entire move to California. And then so m. I'm like this could be the best thing that possibly happened. So that was my approach. I got out there the next day, I started like posting on lin, I started networking. I uh, sort. I interviewed like 10 or 12 different companies at that time and ended up landing ah, at Calix. And I, I mean the layo was end of July and I had my offer by the end of August to join Calix. And so I was able to bank all that money and I was like this is because I didn't just sit there and like you know, weigh to my own self pity. I was able to like put tens of thousands of dollars aside for this move and then start this new job. So that's my approach to stuff. And I understand why people get upset and complain and whatever but it just, I don't know, for me it's like if feeling bad happens, okay, how do we, how do we turn this into. You know, it's the rhyme, you know the, the slowing thing, like the obstacle is the way, you know, it's like how do we, how do we turn the setback into actually something that can benefit us more so than if it's hadn't even happened? I know that a lot of that is like cheesy like woo, woo. But at the same time it's absolutely, I think if I can teach my kid anything it's that, you know, it's, it's things happen. Deal with it.

Speaker A: So before AI like AI became big, like you took uh, a boot camp, you know, one of those tech boot camps. Did that help you with the Pre AI foundation and with the way AI moves, Can somebody skip something like that nowadays?

Speaker B: Yes and yes. So yeah. So when I was at Shopify, I like, I was Recruiting for, you know, engineering leadership, both, uh, tech leadership as well as people management. And. But I didn't know what that meant. You know, I, and I, I'm someone that's, you know, I think this is one of the reasons that I have found my path in whatever way it's been, is that I'm a very curious person and like, I will go down these rabbit holes of like, I have to understand all of this stuff. Probably also adhd. But uh, uh, I didn't know like I was recruiting engineers. I didn't know what that meant. And I'm being on a call with one and being like, hey, can you just like share your screen and show me like what code looks like? Because I don't, I don't even know what that mean. And they, they did. And I was like, oh, yeah, cool. Okay. Means nothing to me, but okay, this is a bunch of like stuff on it. So like, I don't know what I expected it to be like, I, but I was just like, okay, this is, this is the matrix. I don't know what this means, but it did stir this. Like, I want to know what that means. And so I, you know, Shopify had a, like a learning and development budget and so I was like, hey, I want to take this like bootcamp coding class to like learn front end development. And so they were like, awesome. So I did And I learned JavaScript and React and SQL and you know, HTML and CSS and just, you know, general software development, life cycle type things. This was all before AI and I didn't know AI was a thing that was coming. This was 2021 I think, and I loved it. I mean I built tools for the team. I ended up, I started to become like the technical resource. I built onboarding material to help new recruiters coming in that had never done tech recruiting before to understand what the difference is between what is JavaScript versus CSS versus digitimat M. Like, why do we need these three things on the front end? What is backend development? What is infrastructure? What do all these things? And so I was able to use that knowledge to not only build tools, and some of these tools would take me 15 minutes now. It took me three months to actually hand write code from scratch at the time. But it taught me a lot about what that looks like and how to do it and how to approach builds like a software engineer would do. I think people have to do that now? No, but I do think people need to start approaching Vibe coding with a slightly bigger lens than just Vibe Coding, like there also has to be like Vibe product managing and vibe QAing and vibe Security reviewing and vibe PRs and vibe Code reviews. And it's not just like engineers don't just wake up and go to their keyboard and like, I'm going to build OpenAI today and then they go and they do it. That's not how it works. And almost anyone that I've talked to in the past six months, I remind them that most engineers at most tech companies are not building new features, they're maintaining current ones. And so it's not about just building a thing and then it's done. It's about thinking through this entire process, what goes into it, who are the stakeholders, what's the security piece, what needs to be updated, how is this going to be maintained? Who are the stakeholders that need to be brought into this? Who's going to be reviewing the code that I push? Going to be checking to make sure there's so many pieces to this that I think we just forget. So while I don't think it's necessary to understand how to declare a variable in JavaScript, I do think it's helpful for people to get a sense of what things are outside of the small scope of just building and sort of pull back a little bit to understand what the whole life cycle of an app actually looks like.

Speaker A: And this kind of goes into. Next few questions are literally wrapped around that. So Zapier asked you to build out a talent intelligence function and you told me like, your, your honest reaction initially was like, what does that mean? I gotta ask you, like, what did you do on day one of actually building that?

Speaker B: I mean, before I built it, I just asked questions. I think that like, because I, uh, you know, when I, when I would think of, um, the first thing I would do is like, okay, if I'm a recruiter, what would this mean to me? And so I, you know, I put some thoughts down, I had some conversations with Claude about it, blah, blah, blah. But I'm not building this for me. And you know, I'm building this for the 10 plus recruiters we have on our team. What do they want? Like, what's useful to them? So I, I built like, I sent out a simple like Google survey, I put some stuff in a Slack channel and I just asked them like, hey, what would be useful? What would you want go, like in the perfect world, like regardless of how easy or hard or impossible it'd be like going into an intake meeting with a hiring manager, what would you want, uh, going into an offer call with a candidate, what would you want? Going into a sourcing session with some other recruiters, what would you want? Just like, what are things that would be necessary to you? Because there are quite possibly some things that somebody that has done this for a long time might think everybody wants or needs or they've needed in the past, but that's not how this team operates. And so I didn't want to go into with any kind of like preconceived anything. I just wanted to ask them, what would you find useful? Like, how can I help you? That's my job, is to make you effective. And so I took all that information and then I sort of put it all together in terms of, okay, this is all the kinds of stuff they would want. And then I started thinking about, like, well, how, how would they want this information? Like, what would be the most useful way of doing this? And by no means am I saying that, like, this is the right thing for every company. And I think it's always going to be different, which is why you always got to start with, you know, uh, salespeople call it discovery calls, like, what are we trying to actually solve for here? But what I sort of started thinking about was they want information. Nobody told me they want a dashboard, right? Like, nobody said, I want a dashboard that has X, Y and Z reports on it. Or nobody said that they said they wanted information, right? Those are two very different things. One of those things is information. One of those things is a display of that information. Nobody said they wanted a display of information. And so maybe it's just my naivete because I've never done this before, or maybe it's truly a huge benefit of not being in this industry for as long. But I was like, what can I do to get. I bet you there's APIs out there for a lot of this stuff. Let me see what free APIs are available that have this kinds of information. And so I found out that indeed has a hiring lab API, the United States Department of Labor, ONET, Bureau of Labor Statistics, probably four or five more. They all have free APIs. Uh, warn, act. These are all places that have. And I took those and I built an internal MCP that pulls from all of those APIs. It also pulls from Ashby. So we have like full on, uh, current information available to us. It pulls from even some of our internal context. It pulls from a couple of databases we have of candidates and stuff too. And I built this MCP that now allows them to run a cloud skill using this mcp, like if they get a new role and they have the title of the role in a job description or even sort of a job description, they could run a skill and it's going to output really whatever they want. But like the, the demo that I gave them was like, it's going to show you what's the, what compensation, you know, should we be looking at here based on X, Y and Z? Well, here's what startups are offering, here's what, you know, mid size are offering, here's what enterprise companies are offering. Here's a list of 50 companies that are, uh, our competitors. Have any of these companies hired this role in the past six months? Oh, two of them have. What are they charging or what are they paying for that role? Oh, this one's hiring for now. So we're probably going to be like interviewing the same people with them. So we should do some extra research on this company. What's the talent pool in the us? What's talent pool in Canada? All, just all of the information, you know, what's, what was the time to hire last time you hired this role? Okay, cool. So what should we expect this time? Oh, this is one level up, so it's probably going to add X or Y. And so now you can go to this call with the hiring manager and you have this information and maybe you want to build a new dashboard, maybe you want it just pulled into cloud and maybe you want it in a Google Doc, whatever. I just built them a way to pull this information however they want to. And that to me is an effective build because I'm not telling them what they need or how they need it. I'm just giving them access. And I think that is one of the huge areas of leverage that talent, engineers, talent teams, whatever can lean into is data. Like I was saying this, I had a call with a founder yesterday who was asking me about what I do and I was saying that I think that he's like, if you could build anything, money was no issue, like no question, what would you build? I said I would build multiple data pipelines with everything and anything I get my hands on so that when a recruiter needs something, they can get it, they can get it how they want it. They know that it's accurate because it's not cloud scraping Google or exoscraping Google. There's a place for that too. But this is pulling accurate information through APIs onto your computer and then you can do with it what you want to do with it. And so that was How I started the whole process probably took a month to get from, like, what is talent intelligence? To, like, here's the first iteration of this mtp. But I think this is the kind of thing that we should be looking at more of. It's not, can I rebuild a sourcing tool or an etf? Yes. It's what are. How can we get data in their hands? Because if you have a good recruiting team and you give them good data, they're gonna do a really good job with hiring.

Speaker A: So you've said, you've, uh, said this. Like, most people right now are building. What they're building is great for learning, but a lot of it's not really moving the needle. Like, where is that, that true line in your head? Like, where do you think that line is?

Speaker B: I think it depends. I think it depends on the company. I think it depends on the person. I think it depends on a ton of things. I think it depends on how that build got started. Like, did this build get started because someone was just like, let me see if I can build this thing that I think is useful? Or did the build get started because they did some research and did some interviews and talked to some people and decided what was the groundwork done before this bill happened? I think that one of the easy targets is sourcing. For myriad reasons, I understand why people would want to get into that space. And I've actually talked to some people that have done some really interesting and cool stuff in building in the sourcing space. And I don't want to blanket say that that's not what we should be focusing on, but I do think that there are other things that will move the needle more than the amount of time it might take to do that. Anything could move the needle depending on the company. But are we checking that? Are we running the data? Are we pulling reports from before and after this tool is in use? What is the name on this tool? Like, there's so many things that go into that, determining whether it actually made sense. But the way that I approached it is like, look, if you're learning to do this, I don't care what you build. Build something that exists already. That's probably the best thing to do, actually, is build something that exists already. Because then you can see, you can compare. It's like, oh, cool. My tool does this. Oh, there' this. How do they do that? Let me try to do that too. And then you go and you try to figure it out. That's super cool. But building to learn and building, because you truly Want to ship this thing and impact the data or impact hiring are two very different things. And they're different projects start to finish. Like the way you start it, the way that you operate it, who you work with is all very different. Learning can be done by yourself. True projects most of the time are going to need at least one other person. Like, you should never build in a silo, right? That's the whole reason that code reviews and pull requests exist for engineers, right? Like no one just, I'm going to update this, uh, you know, the Shopify database, and I'm just going to push the code and it's going to be good. And it's like, oh, I broke Shopify. It's like, well, yeah, don't do that. Right? Like, that's why you have to have at least one other person reviewing your code before it goes in. You have to find ways to be able to like, branch it off and pull it back just in case you did do something. And I think that, you know, again, this goes back to an engineering mindset rather than just a builder's mindset. But, uh, I'm all for people building, but when they're building for a purpose, we need to approach it in a little bit of a different way if we want to see some kind of ROI on it.

Speaker A: So companies right now in the recruiting space, recruiting tech companies are pouring tens, if not hundreds of millions of dollars into this. Juicebox just raised $80 million at almost a billion dollar valuation. PIN, uh, is a sourcing tool that I use. They just kick out agents. Ashby's kicking, uh, out MCP support later on. Uh, they've also shipped agents. Atlas, who's a big sponsor of this podcast, works with cloud. I can literally just hop in cloud and start asking stuff and it completely taps my database. So, like, what is actually left for like one person on a talent team as a recruiting engineer to actually build?

Speaker B: Yeah, it's a great question. And I think this is the reason that so many folks kind of default to. I'm, uh, going to build a sourcing tool because one, those costs a lot of money. Um, and two, we're sort of, we're looking at things through these big buckets. Like, well, I have my ats, I have my, my sourcing tool. I have, you know, and it, and that's fine. And again, if you want to go through the process of building out a tool that you think will replace one of those things, awesome. Do it properly. But realize more than one person works at Juicebox, right? It's not like, and it's not like they just pushed it out and then it's done and then it just runs. And we're like, cool. Like, if you're going to build something that's comparable to what they're doing, realize it's going to be more than one person and there's going to be quite regular maintenance on this thing forever, Forever and ever. And if you have budgets to do that, cool. Like, I forecasted out some, because we're trialing some tools right now, PIM being one of them. And when I forecasted at, like, the cost of builds, of hosting, of maintenance, because something like that, I don't. I wouldn't want just to be in my hands. Like, I would need to partner with engineering. And so now we're pulling people over to spend their time building something for. It's. It gets very expensive very quickly. And it's like, is that really more expensive? Like, pin charge is $179 a month per seat, and you can just turn it off the next month if you don't. Like, is it really worth that time? Maybe it is. And I don't know. Right? Like. And I think. But ask the question. Like, you gotta. And one of my favorite podcasts is, uh, it's called Money for Couples Ramit Sethi. And he, uh, says all the time, he's like, run the numbers. You've gotta run the numbers. You know, and if you're buying a house, don't do it on vibes. You gotta run the numbers. You know, are you spending too much on your. On your rent? Run the numbers. Are you, you know, are you putting this enough aside for retirement? Run the number. You gotta run the numbers. Because maybe it does make sense. I don't know. But we're making these decisions on vibes so much, hence the word vibe coding. And I just. I. So, anyway, I know that didn't answer your question, but I think what that leads to then is, hey, if these buckets are taken care of, to your point, like, what. Where can we move the needle? Right? And I think this is the other part to, like, what makes a truly good talent team is let's rethink stuff. We have these buckets. One, are they necessary at all? Do we need this? Two, do we need application forms? Do we need screens? Do we need, uh, interviews? Do we need a careers page? Do we need, like. And maybe the answer is yes to all those things. Maybe it's notice some of them. I don't know. But, like, let's ask questions about this stuff. Like, how can we make candidate experience? How could we 100x candidate experience, right? The great book called Unreasonable Hospitality that talks about just going to the moon for every single customer at your restaurant. How could we apply that to recruiting? Like, every single person that applied for a job walked away thinking like, what an amazing experience. Like, how do we do that? Right? Because that's not happening right now. No one's getting that experience. And so it's like, it's rethinking these things. It's rethinking things like data pipelines, it's rethinking ways that we close candidates. It's just questioning everything because now we have tools we've never had before. And so it truly is like, let's ask questions about everything that we've taken for granted up until this point. And maybe everything still stays, maybe nothing stays. But more than likely it's probably going to be a combination of those things. Some things still make sense and some things actually don't in this landscape. So if they don't, then what does that mean for us? And how can we make this better? How can we stand out? How can we make that, uh, candidate's experience special, even if they're one of a thousand people applying for this job,

Speaker A: standing on the thread of uh, moving the needle. It's a trap that I've seen myself and a trap that you actually described to me. You know, you're supposed to be saving time with AI, but with that time saved that you're supposed to be saving on, um, AI, I sometimes catch myself reinvesting, ah, time back into cloud or back into lovable. What should people be doing instead with this technical time saving that we keep on talking about?

Speaker B: I think again, it depends on sort of what your, what your ultimate goal is, what your role is, all that stuff. But for me, like, uh, recruiting teams, if AI is being leveraged in a really strong way, it's going to give recruiters more time to spend with candidates. It shouldn't lead to less time, whether that means they're doing more screens, whether that means they're doing less screens, but they're spending more time later in the funnel having conversations or having preparation calls with candidates, or having follow ups after interviews. Where can we add these touch points to make candidates again to make that candidate experience. I think we all think about recruiting right now with AI through the lens of like, how can we save recruiters time? How can we get faster? How can we make art? And I'm still struggling with like, well, what if we don't care about this. What if we like, only look at the candidate? How do we make the candidate better? Like, how do I make everything on that side better? Right? Like, how do we make their experience better? Even if our numbers get slower for a while? Like, what if we could make that better? And so I just think that's the point of all this. And like what we started with where it's like, uh, this is a people job. And if we're using AI to get further from people, I think we've lost the forest for the trees. Trees. Like we're just, we're roaming around and we're going down this path that's like, I don't ever want to see top of funnel completely replaced by AI. Uh, I think that AI screens absolutely have a place. I think they're incredibly helpful and I think they actually benefit candidates more than they benefit companies. But I still don't think it's a replacement for that initial recruiter call, for that relationship building. Because we're closing a candidate from the very first conversation, you know, and people forget that, right? Like, we're not, yes, we're assessing them, but like, it's a two way street where we want to make sure they are getting a really good impression of us. And like, if I'm, you know, when I was recruiting and if I was closing a candidate, I would want to look back at that first conversation I had with them and be like, if they were basing this offer on my first interaction with them, would they accept this or would they probably walk away? Was I kind of like a jerk, uh, to them? Was I kind of like, I'm the gatekeeper here, you got to impress me before we move on. Or was I like, truly like, like treating them like a human? They're treating me like a human and we're like, let's see if this is a fit. Because we're always closing people, like first call, last call, doesn't matter. And so if we forget that and we start outsourcing to AI, that closing call gets a lot harder. And again, we've just lost, we've lost the plot at that point.

Speaker A: So, uh, I'll just say this, like, I'm building something that's probably totally ego build and lovable and you'll probably laugh at, you know. And I know a lot of other people are building too. Like, I actually am overwhelmed with the amount of like, hey, can you check this out? What I'm building conversations I'm getting now, you know, with where you're sitting and what you're doing. Is there a version of these builds that you're like, okay, that's no longer ego build. That's like a version that I respect.

Speaker B: What does that look like when it works? I mean, that's the. That's the answer, you know, like. Like, I think it's. If you've built something, it almost. It's not even the build. I think it's just the way that it's presented and when it's presented, like, if you build something today and post it online tomorrow, that you've just changed the landscape of recruiting with this thing. Okay, but based on what vibes, again, like, you just. You think it's going to make. And I. Someone, actually someone that I, that I hired at Shopify posted this on a version of this on LinkedIn yesterday. But they're basically saying, you know, to be an effective builder, like, you can't get too attached to your ideas. You may think it's the greatest thing in the world, and maybe it is, but maybe it's not. And if you don't have that ability to detach from ideas early, when you're getting feedback that it's like, no one wants this. And it's like, why are you doing it? So I think that if you've built something that you've deployed, you've shipped, uh, it's gone through whatever SDLC process it needs to go through. Adoption happens. And the metrics are showing that this is a tool that's like, truly, not to use the term yet again, but moved the needle. Awesome. I'm, um, thrilled. Like, now tell us all how you did it. Like, I would love to learn from you because it sounds like that's amazing, but it's. We won't know that for a while. People are just building these things that aren't going to. I can't even imagine the amount of tech debt that exists in the world now because of people's, you know, projects that have failed. And again, great. Like, I want that. I have way more failed projects than I have successful ones. But what I'm not doing is like going on LinkedIn and telling everybody about this build that I just did that no one's ever going to use or touch that doesn't actually help anybody. So fail. Awesome. Build stuff for sure. But maybe just wait a little bit before you tell the world how game changing it is until you've actually shown that it's game changing instead of just that kind of thing. So maybe it's ego sharing of builds rather than ego Builds.

Speaker A: But yeah, uh, here's a question more for the agency side of the house, the boutique firms, the small firms or even the solo shops, you know, a two to person firm. What is an agency version of like this look like? What is something that them or even a solo recruiter can stand up over the weekend that can really just change their next client pitch?

Speaker B: I mean I think it like I'll sound like a broken record. This is at least right now this is the hill I'll die on is like it's the data piece, right? Like build yourself an mcp. Build yourself the ability to pull information when you need it to be able to get information on. Companies use clay to have enrichment set up webhooks so that when things change you can, you know, in the moment instead of finding out like a week later, subscribe to Warren. Like even like half these things just have emails you can get on like daily basis. Like Warren sends an email on a weekly basis that lists layoffs. Like most recruiters don't even have that. Like these are real, real easy things. But for me it's like if I was on the agent side I would want to arm myself with as much data as possible and then use that different ways depending on the client that I'm working with or what they're looking to do. I would absolutely want to make sure I'm staying up to date on all of the more so than I currently am. Like um, I'm a lot of people are always obsessed with oh you know, Claude just dropped opus this and you know, and I get it. But I'm also like that that's not as important to me. Whereas if I think if I was on the agency side I would want to stay a little more up to date on every new release that each of these providers was releasing and I'd be constantly testing things out and sort of seeing what this model does versus that model is it pull better information or whatever. But uh, you know, for me the first thing I do is I build out the Talent Intelligence FCP and just start there with that information and then prove that concept to yourself. Because it's not you could build out on a weekend and it's again it's all free. Like it's not. I mean obviously tokens to build it cost whatever. But like the APIs themselves are all free, all the endpoints are free. So it's like this is a really low cost tool you can build for yourself and then start to use it and see how it's helping and see what you wish you could do differently and then start to build those things. Know you learn so much when you start to use something you've built, you learn maybe you want a little bit differently this way. You want, you wish you could do that thing or, you know, I have this information but man, if, like, if I had this piece that would really change the game at this stage. And then you start to just go down these rabbit holes of discovery and like, oh, this is super cool. Now I know that when I kick off, you know something with this, I can run this skill that's built and, and you just, it just snowballs into like this wonderful sort of, uh, data sciencey sort of approach to recruiting. So I think that's where that's where I would start, is just like how much knowledge, how much info, how much data can I pull in?

Speaker A: So something interesting has happened at least in the ATS and side of the house over the, over the last few years because I remember like how clunky many of the ATSs are. And now and Atlas is a sponsor, I use Atlas myself, it hooks in with Cloud. So I can literally just go into Claude and ask like what's happening with these candidates or where are things at? And PIN is now in Claude too. So I can just like, hey, source with pin, work with Atlas. All this stuff is just happening. There's like, I almost feel like there's two futures with an ats. One where the ATS literally has everything inside it, including like an LLM or was just headless, where it's there but you never actually touch it. Which one do you think we're going to be living in by 2028? And what happens to the recruiters using H?

Speaker B: Yeah, I don't. So 100% agree. And actually there's a, I think there's a middle ground there too, of options, which is you, you're using a headless ats, but either through that LLM or through claude, there's actually code built into it that allows you to build custom UIs. And so you're still using it headless, but maybe you're building your own UI in, you know, on your local host that allows you to sort of operate everything you need. But, uh, you're like, you can put your roles there, you can add a calendar here, you can have your LLM chat window at the bottom, you can have candidate feed coming down the left side, like all this kind of stuff. So I think there's, there's sort of uh, a middle ground as well. I don't think we're ever going to see one or the other. Like, this is a black and white conversation. I think that if ATS is like, move towards that, where you truly can operate it, like, everything can be done from within Claude or within your terminal. I certainly think that there will be people that will adopt that and will live that way. Only going in if they really, really have to, like, to whatever UA they may have. Will everybody do that? Absolutely not. Like, there will always be a need for. It's easy, like in the tech bubble to forget that, like, not everybody is building cloud skills and using custom mcps, right? Like this. Like, people. Some people have just today probably searched on Gemini for the first ever time and they're like, whoa, this is crazy. This is different than Google. What? This is AI. And so I think that we're always going to have, like, people are not going to want to constantly be adapting to new things that are changing all the time. They're going to want to have something that's comfortable. And I totally get that. And honestly, like, I think that as I get closer to the end of my career, I'm sure there'll be a time where I'm like, I can't just keep learning, y'. All. Like, this is just too much. Like, I need to just, just be happy with one thing. So I do think that there's. I think both will exist, but I do think that the overall point is that connectivity is used to be something that we thought was a pipe dream. And now if, like any vendor that I talked about, any tool that we might work with at Zapier, and you can ask everyone that I've talked to, all vendors, one of the very first questions I ask is, hey, do you guys have an mcp or do you have an open API that I can build on? Because if they don't, it's never going to be black or white. But if they don't, it's like, ooh, why not? Like, is it in production? Like, are you hoping to ship it soon? Because, like, I try not to leave my terminal. So if there's a way that this is a thing, or maybe you want to like, test it on somebody, like, send it over to me. I'd love to give this a try. I really do think that that's the, that's the way of the future for those that are tech inclined is truly living in one place, um, maybe building their own UIs for that information. But everything is pulling in accurately. And again, we're in such early days of what this looks like it's going to be, it's going to be night and day in 12 months from now. What MC and I'm already talking to developers and I was. Someone in Promptmates was saying yesterday that like, developers are already like, I don't use MCPS anymore. It's just, I just write scripts for APIs and so they're already moving past MCPs and they're, you know, which I get as well. You know, like I remind people this sometimes, like APIs existed before AI. We actually had ways to access APIs before we had Claude. And I think people forget that and they just use it for everything. And it's like, no, like if you're pulling the same information all the time, like, and you have API, don't use Claude to do that. That's costing you tokens every time. Like just have Claude write you a script that uh, when you trigger it pulls the information you want every single time the same way and you're never paying for tokens again. So it's like, how will that evolve? How will MCP evolve for those that are truly tech inclined? Will we actually move away from mcps? A month ago, loops were all the thing everyone was. And I saw a podcast of engineers saying loops are the only way we're going to code from now on. Uh, my loops have loops and everything is loop, loop. And then a couple days ago I read a post that's like, yeah, loops are so over. No one's looping anymore. And it's just like, man, this is like a 30 day life cycle for this. Like, so I, I get it. It's gonna look so different in like 12 months and in five years. But yeah, I uh, think that we just, we need to be adaptable and we need to be able to do, we need to be able to work whether it's in the ui, whether it's in Claude, um, and just be flexible and always, but always pushing those boundaries, always like asking for things that aren't aren't there that we want.

Speaker A: With how fast things change, do I have to like drop this next week instead of like two months like it was gonna be?

Speaker B: Yeah, they'd be like, oh, Claude was still around when this was, when this was recorded?

Speaker A: Yeah, uh, pretty much.

Speaker B: Oh man, Rip.

Speaker A: So, so interesting thing happened in sales like a decade ago. It, uh, used to be a salesperson did a sales thing and now you have fast forward almost one salesperson, almost has one in sales, engineer person behind the scenes to kind of like do everything. Are we going to Start seeing the same thing in recruiting where you have like a recruiter, like a 3 to 1 ratio with a recruiter versus a talent engineer or recruiting engineer.

Speaker B: Yeah, I love this and this is, this has been, this has been my thought I think partially because I let all the pre sales hiring at Shopify for about a year and a half. So I hired and built probably half of the solutions team at Shopify. And I don't know if I've ever loved hiring managers and org as much as I love of the solutions, engineering and architecture teams at Shopify. I learned so much from them. I did not know what a solutions engineer was. Again just similar to me when I started recruiting for engineers. When they put me on the solution side I was like I don't know what this means but it truly it clicked something where I was like oh this makes so much sense. And as the talent engineering thing came I was like that's what this is going to be. Not quite the same but I think we're going to start to see this relationship and it might not be like uh, on solutions teams like depending on the, on the verticals like enterprise Shopify anyway is like a one to one relationships. Like you truly have like a partner there. Um, SMB could be 20 to 1. I've talked to people that are, that have 50 AES that they support. As a solutions person I don't know what that ratio might look like or what it might be but I do think that companies that are like forward thinking about this and are putting that understand the value that someone can bring when they're building things. I do think that model is something that could work at uh, companies now. I don't know, maybe that's totally wrong and it would be a disaster and I don't know that either. But I do think that something like that would be worth exploring as ah, an avenue for how this role evolves because then you have the people you partner with. You're not trying to build for an entire team that would be doing things differently. You're the talent engineer for the engineering recruiters and so you have these seven recruiters that all recruit for engineering and then you get to approach you're building through that lens. You don't have to worry about how the commercial side is working or how salespeople or what their process like. You truly get to work in this lane and I think it opens up a ton of possibilities rather than having to sort of tailor builds for an entire org of people that work quite differently. I think there's benefit there too, 100%, don't get me wrong. But the alignment in verticals opens up a lot of doors.

Speaker A: Just with everything that you know about the engineering behind the scenes for recruiters, could a three person agency do between 5 and $10 million a year? If one of these, one of the three is a builder, could they?

Speaker B: Yes. Is it likely yet? Probably not. I think in order to do that you really have to be dialed in with what you need, with what has been successful for you thus far. And how do you scale that using technology?

Speaker A: Right.

Speaker B: It's like let's look at the things that have gotten us to the point where we're currently at and if we want to get to this, this point here, what does that look like? Like what are the things that we would need regardless of how we get them, Whether it's through technology, through people, through whatever that looks like. Like, let's just start with the end in mind, right? You know, covey seven habits, highly effective people. How do we get here? And let's work our way backwards. And then not only do you have to kind of map that out really effectively and determine if it is even realistic, period, but then you need to have somebody in the technical seat that understands building, not just someone that wants to jump into cloud code and like vibe code, a website real quick. You know, it's like it's someone that needs to go slow before they go fast. It's someone that needs to ask all the right questions. It's someone that understands, I mean this is basic but like understands what GitHub is and like how to, how to use the tools that are out there. Like, and if, I think if you have all of the ingredients right, you could probably do something that is not yet been done in the agency side. But I don't know if we're there yet from a knowledge standpoint, myself included. So yeah, do I think it's possible? Yes. Do I think it's going to be done in the next six or 12 months? Probably not, but I do think the sky's the limit if you're using things properly.

Speaker A: So this is a brand, like the recruiting talent engineers, like literally a brand new field, a brand new role. Where do you see it in five years?

Speaker B: I ask myself this question all the time. And when I took the role here and left recruiting, a conversation I had with my wife was like, this is a kind of a risk, like I don't know what this role is going to turn into, but I suppose if this doesn't work out, I can Always. I'm a great recruiter too. I can move back into that space. I'm not like I'm having just then a complete 180. But, uh, that was the question is like, do I see this as something that is going to be long term changing tech or is this a flash in the pan that we're going to go through this building stuff and then we're going to just like loops like, oh, talent engineers. Okay. Yeah, 2026 called, they want their talent engineer back. And so, I don't know, I think that, I do think that it will persist in some way, shape or form. I actually personally think that if people start to think about this in the right ways, I do think this will shape the industry. I think if we get people to just want to vibe code stuff for people and that's what they think talent engineering is, it's going to sour the market. It's going to poison the, well, whatever metaphor you want to use. But people are going to be like, oh, no, I tried to hire a talent engineer and just all they wanted to do was vibe code 27 apps for me and send me 300 dashboards. And like, I don't think this is like a useful, like function at our company. It's like, no, that isn't. But I, I don't think that's what talent engineering is. So I think that if you have people that are actually thinking about this in the way that engineers would think about this, and that's why I like the A16Z thing is they actually think about it more. Like if you're an engineer who likes to recruit versus a recruiter that likes to engineer because you have to have some of those principles in place. And if you don't, I think you're going to, you're just going to want to do the wrong things because it's fun. And you know, I'm not saying it's not fun sometimes, but like any engineer, I'm not calling myself an engineer. Like talent engineer is very different than software engineer for sure. But it's not always fun for everybody. Like as soon as you're, you know, literally in there, just like making the small, you're like, I know I added a comma somewhere. Where is that extra comma? Because it'd be 30 minutes to find this comma and delete it. Oh. So, you know, it's not always, it's not always fun. But I think if you have the right people thinking about it the right way, I do think this has the capability to Shift the way that talent teams work sort of permanently moving forward. I hope that it does. I hope that we have smart people in place to be able to do that. And that's again, why I'm excited about the fellowship, is because I think there's a lot of smart people in that group and I have a lot of strong opinions loosely held, and I feel like they will too. And I can't wait to have conversations with them and see what they're thinking about.

Speaker A: So if somebody's listening to this, whether they're agency or even in house, and they're like, hey, I want to become like a recruiting engineer. I want to become a talent engineer. How do you actually become one?

Speaker B: I think the default answer that I see everyone giving is like, just build. Like, just go out there and build, build. You know, And I agree with that. I will say that. I will also say that I don't think that is helpful for everybody to hear because what does that mean if you've only ever used AI? Uh, because you have copilot at work and you just have it to ask questions to about a company context and it points you to a Microsoft page. If I say build, you're like, what are you talking about? What do you mean build with AI? That doesn't mean anything to me whatsoever. And so I like to give people a little more, I guess, of an example or whatever. But, like, I want you to go to Claude and I want you like, if you're truly someone that, like, when someone says build, you're like, just the, uh, world goes black. And you're like, I don't know what that means. Like, stop talking to me. Go into Claude and tell it you want to build a really easy local host, local locally hosted webpage. And example I've been giving people lately is that like, my best friend lives in Toronto. His name is Scott. He's coming out to visit me next week. And I just, I built like a, a fun landing page that's hosted on Vercel that has his flight information, it has some photos of us, it has some fun facts about Folsom where I live. It has some, you know, it has a. I pulled in a weather API, so it has like the forecast going to be showing up for the days that he's here, that'll update regularly. And I was like, this is. This did not take me long. Even has like a plane that flies across the top. Uh, this is like, catch me, idiot. If you move your mouse towards it, it like runs away from the mouse. So you can't, you know? Totally. Like in no way. Like this is not a, this is not a useful business tool. But the point is, is that, like build something like that. Like just tell Claude, hey, I want to, my friend's coming to visit. I want to build a, just a fun landing page for him that has some basic information about the trip. Like maybe things he should bring that he hasn't thought about, weather, whatever. And then, and just see what happens. Like, like have it ask you questions and then move through and build it. Maybe don't include weather at first, just include like a very static page that's hosted on your computer that you can just open. It's just HTML. That's it, Just be super basic. And then if you spend a couple hours in that and it's working, be like, this was super cool. And then you could be like, hey Claude, if I wanted to add in live weather, how do I do that? What is an API? What does that mean? And let it walk you through it. Start the whole thing by saying, I've never written a line of code in my life. I don't know anything about technology. When you talk to me, I want you to talk to me like you are Talking to a 12 year old and that's it. Like truly, you know. And if I want you to go deeper, I'll ask you, but like don't use technical terms. Or if you do give me an explanation, or if you're someone that likes metaphors, ask it to give you metaphors. Or if you're someone that likes visual descriptions, ask it to whatever. Like find how you learn. And then as it's going through things and being technical, just have IT explain that stuff to you. But something as simple as that is a really good way just to like build it so you know, and no one else can see it because localhost is not shareable, you know, and if you want to start to deploy it, have that conversation too. Like, hey, if I want my friend in Toronto to see this, how the heck do I do that? And again, explain this to me like I'm four, you know, and just walk through that process. That's what build means. Like just make something on your own computer that no one else can see. Don't worry about anyone else seeing it, that's a whole other build. But just do that and then if that works, add stuff and then maybe add some more stuff, and then maybe add some more stuff. And then only other thing I'll tell people is like, your first 10 builds are going to suck Most things don't work. And that's how it should be. Like if everything you build is what you ship, you either have a real low bar or you're really, really good at what you do. So that's how I want people to start. It's just like, truly, there's no stupid questions, there's no dumb thing to make. Just have fun, have it build you a learning module on how to build, you know, like, have it walk through you and then you can like, like have it when you're done and you can actually go back and look at the things you learned in this thing that you built on how to learn it, you know, like, why not? It's great at that and ah, it's

Speaker A: crazy how fast and better it's gotten. I think it was like last year I was trying to build out the Builders Club for the elite recruiter community. And like, I think it was like 75 screenshots later. Claude, give me another screenshot. I'm like, I don't know what I'm doing. I'm like, talk to me like I'm a five year old. I finally ended up just hiring a software engineer to build it.

Speaker B: Whereas today it'd be a whole different story. Like you just, it's, yeah, it's amazing. But like, if you treat it like a person relatively, I mean it's not perfect and I don't want anyone to think that it's always going to be right either. And that's a, uh, whole other conversation we won't get into today. But like, for stuff like this, it's very good at walking you through how to build something.

Speaker A: So before we jump over to the quickfire questions, is there anything else that I should have asked you or you want to go deeper on?

Speaker B: No, I don't think, really. I think I just, I want people to explore this. Like, if, if this is something that's interesting to like, I, I think the, the mentality, and this is kind of similar to the question we just talked about. But like I, I don't like the mentality of like I'm too far behind or I'm not. What, like in the grand scheme of things, like this is really like day one or two, maybe we're towards the end of day one, but like we're not even a week into this, let alone whatever. Like AI is brand new in the world and nobody is behind. And you may feel behind because of Insta Life and LinkedIn, Ego Build shares and stuff like that, but like you're not. And there's no bad time to start learning this stuff. There's no bad way to learn it. Like just jump in and do it. Find people that are of similar minds. Like uh, I mean I'm in the prom mates community and it's, it's amazing to talk to people that are in that and like find out what they're doing and just I get to share some things and with them and help them. And I've learned just as much as I've taught other people in that group. And it's just surrounding yourself with like putting yourself in the environment of people that want to get better, whether they're better than you, whether they're newer on their journey. You're going to learn stuff from all those people. It doesn't matter. And so just stop the like fear or whatever, uh, that you think you have of. You know, I don't know this stuff and people are already way for. Like there's no point in me starting at this point or I don't, I don't even know.

Speaker A: It's.

Speaker B: Don't be that person. If you want to do it. If you don't want to, fine. I'm not pressuring you to do it either. But I think that if this thing you want to do, there's really no reason not to do it. It's going to be hard, but literally that's how you learn anything. If something's not hard, you're not learning.

Speaker A: Jump into the quick fire questions. They don't need to be quick answers. But what's one thing that you believed about AI within recruiting 12 months ago that you now think is flat wrong?

Speaker B: I think something that I've changed my mind about in uh, recruiting in AI is I used to be incredibly against AI screens for interviews for like initial stage. I was always like, recruiting is a people job. If you're sending people to an AI instead of a person, you're doing it wrong. And now I'm really like over the past probably six to eight months, I've realized it's a much more nuanced conversation than that. I think there are very good ways to use tools like that. We use Ezra at Ah, uh, Zapier. Um, and I think there are bad ways to use tools like that. If you're using it to truly replace the recruiter at top of funnel, I think that's the bad way to do it. Like if you have three applications and you move five people forward to the recruiter screen but it's an AI, I think you're doing it wrong. That's my opinion. And you can disagree if you want to. If that's how your company does it, awesome. Good for you. If you're seeing amazing results, cool. Like, I won't. But for me, that's like, I don't. Again, like, if you're closing a candidate, I want to look back to that first conversation and see how I did. And if we're skipping that, I think we've already again, lost the plot. I think if you use it as more of a. Either part of the application process where most, if not everybody moves to that stage, and now it gives candidates a chance to actually share context about themselves that would never be shared in a resume or an application form. I think that's an incredibly positive use for it. And like I said before, I think it's actually more useful to candidates than it is to companies. It's not saving companies a lot of time right now, like, at all. Because if you get 300 Ezra screens, guess what you're doing as a recruiter today? You're going through 300 transcripts and, like, notes on those. Like, we're not. Because we're not letting that make the decision. It can't. It's not. We're not there yet. Maybe we will be something, I don't know. But we're not right now. That's what I know. And so it's not deciding anything. People are still looking at these transcripts and these summaries and all this stuff and making those decisions for themselves. So it's not saving us time. It's just enabling us to find candidates that align regardless of whether their resume is a perfect. You know, and this is so important to me as someone that was a CrossFit gym owner and then became a tech like somebody at Shopify. Saw Katie Slater interviewed me. Props to Katie. I'll always be thankful for her. Saw my resume was like, oh, yeah, he's been. And teaching muscle ups and snatches in Brooklyn. He should be high, you know? But they saw the transferable skills, and they were right. I bled the entire engineering recruiting team in hiring my very first quarter. And so it's like, you have to be able to find ways to pull people because if you lose that ability to hire people outside of the quote, unquote, like, perfect profile, you've. You're gonna like, again, you just lost the plot. That's not what AI is for. So I think that that's something that I've very much changed my mind on, is it can be used very effectively. It can be Used bad, but it can be used very effectively.

Speaker A: Favorite book that's had an impact on your career.

Speaker B: I have three books that I recommend that every single person in the world read. And I have a fourth book that has been personally and that I recommend to half people that I talk to. Mindset by Carol Dweck should be required reading for every single person and in the world. And if you don't truly understand the difference between a fixed mindset and a growth mindset, I think you are doomed to a life of less happiness. Obstacles the Way by Ryan Holiday. Again, I don't think I would have moved through the layoffs that I've had through my career, moved through the setbacks, everything, had I not had the perspective that I had from understanding more about slope philosophy and just that approach. And I've read a ton more books both of his and of just meditations and whatever. And the third one is Seven Habits of Highly Effective People by Stephen Covey. I think if humans came with an instruction manual, that would be the instruction manual. And if people haven't read that book, stop what you're doing. Because I would put that at number one. And then Mindset and then obstacles the way I think those are just every single person in the world should read them. And then there's a book, uh, uh, Range by David Epstein which makes the case for how important generalists are in the world. You don't need to be someone that's been in a field for 25 years and to be effective. In fact, you can be more effective if you're not that person. You need both. You need to have these deep specialists that have been doing this thing for 30 years and you need someone with a fresh perspective that's not caught in the same patterns that you are. Because no matter how self aware you are, you're still going to be caught in certain patterns and certain. And so being able, that was a huge help in my pulling myself out of the imposter syndrome of being 40 and getting into tech for the first time. Because I realized that, oh, so yeah, these people that I'm working with have been recruiting forever. None of these people have worked 100 hour weeks to get a business off the ground and know what it takes to go through the hell of starting a business. Like I do what it takes to like give everything in your soul to a business, to hire people to see the transformation like and just it allowed me to be like, oh cool, they have a lot of stuff, I have different stuff. We are both, we both need to be here to make this whole thing work. So Those are the four books.

Speaker A: The next question. You can't say LinkedIn, OpenAI or Claude. What is your favorite tech tool at the moment?

Speaker B: Uh, can't say LinkedIn, Claude or uh. So like no LLMs and stuff. Let's see. You know what, you know what? I would find it really hard to live without and I just actually posted about a couple days ago, text expander. This is a tool that Shopify had when I first started working there. And I mean it's, it's simply what that's, it's what it does, it expands text, right? You just have a snippet and you do it. But you can build all kinds of stuff into it. You can build, you can write JavaScript into it. You can do all kinds of like very cool build. Like you can take things that are on your copy sling and you can have that paste into this formula that it output. And so it's a very cool tool even for basic things like I don't type my email anymore, I don't type my name anymore, I don't type my booking link anymore. Everything has like a two or three button shortcut and it's not saving me an hour here, an hour there, it's saving me five seconds here and five seconds there. But like 500 times a day. So it's not even a. I, uh, guess it's not even really an AI tool. It's just a, it's just a tech tool and I think they're moving towards a lot of AI stuff over there. But like I'm just, I like after I got that from Shopify, the first thing it was like paid for your subscription to Texas man. Because I was like, I will happily pay this, like this is something that I don't, I don't want to type all these emails or like even just, you know, my address, my phone number, like when you're applying for jobs and you have to like be pasting your resume and like I want to be able to just have all of the sections of my resume copied so I can just like Texas man it into their application form. Like it's just easier. So I think that, I think the only answer that I give for that, that I'm really, I really dig what they're doing because we use it at, I use it as app use granola. I think it's a very cool AI tool for transcribing and taking note meetings. And recently uh, they updated it so that it actually pulls from, if you connect it Google Calendar and Slack, whatever. Every meeting you go to, when granola pops up, it actually gives you context about who this person is, what the call's about, what the last call was about, anything that's happened since then. And it's just like, this is so cool. Like, if you're back to back in meetings and you just have always this. Like, every time a new beating happens, you have the summary and you're like, oh, that's right. We did talk about this last time, and you're ready to go. I just think they're doing some really cool stuff over there. So that's. Yeah.

Speaker A: So you spoke at the AI Summit that we had in July. It's on your page that you're a talent engineer. You're one of the few out there. I'm sure you get hit up by questions all the time, like, how do you become one? What about this? What do you think about this? Hey, can you check out this tool that I built? Is there, like, all these questions that you get? Is there ever a question that you wish somebody would ask you, but they never do?

Speaker B: Is there a question I wish somebody would ask me, but they never do?

Speaker A: Um.

Speaker B: That's a really good question. I have a lot of conversations with people. I will say, while I think in the back of my head about how to answer this question, that's actually one of the things that I'm. I. I need to get better at is, like, I say yes to a lot of conversations with people, and I need to be a little more protective of my time. But I just. I truly love people, and I want to. Being a recruiter, that's not recruiting anymore. Like, I don't interview. I don't talk to people all the time. So I'm just like, well, this is a cool opportunity to talk to somebody. So I definitely do that a lot. From a purely selfish point of view, I really liked the question yesterday, and I feel like, if you could build anything in the world, what would you build? No one really asked that. I think that's a really cool question to ask because I think gives you. It gives you insight into your own ideas and thoughts that you probably don't ever let yourself ask yourself that because, you know, there are confines. And I think it's really good to pull those confines off sometimes and be like, okay, but if I didn't have any budget constraints, if I didn't have, like, what. What would that thing be that I build? Because maybe you don't get to build that thing, but maybe there's, like, A stripped back version that you can do. And so I really liked that question. I don't think I would have have answered with that before because I wouldn't have thought about it. And it's not even like it's the most earth shattering question in the world, but I liked it and made me think of like, I just never thought about that before. So. I love when people ask me things that I guess the answer to your question is I love when people ask me things that have never been asked before. I like unique questions and I get not everybody and not every question should be that way either. But like, uh, I still think back to one of the best questions I was ever asked as a recruiter. When someone, you know, people ask always about like, so what are the company values? Like, you know, what's, what's the value of the company you, you align with the most? Like those are good questions. But I, uh, had someone ask me once, like, what's one of your own personal values that you always try to bring into work with you no matter what, and hold that value? True. Regardless. I was like, oh, what a good question. I've never been asked that. Like it's always about company value, but they're like, what's something you personally hold as a value that you will bring into work every day and try to hold onto? And I was like, that's really good. Because what a great question for a candidate though too, right? Because it's like that's going to give them so much insight into not only how I think but like how the company thinks through. Like, uh, there's just so many, so much great info to take from asking that question. So I still think with that question to this day, like, what a great question if you're a candidate to ask. So anyway, I know that wasn't the question, but I, yeah, there's a couple of answers for it in reality.

Speaker A: I got two more questions before I let you go. First of all, if somebody wants to follow you or connect with you, how do they go about doing that?

Speaker B: Yeah, LinkedIn is the best way to do it. I am, um, yeah, it's just, I think it's. I am Dan McCarthy I think is my LinkedIn right now. I have added just for fun a little. My name is now Dan McCarthy trademark. I've added the TM just for fun because Slack won't let me do it internally. So I'm just gonna do it at LinkedIn externally for now. But yeah, come follow me on, on LinkedIn. Um, I do have an Instagram and a Facebook that I haven't looked at in almost a year. I've tried. I try to avoid social media as much as I can, but, uh, so, yeah, LinkedIn is the best. And if you want to actually like, like, chat, join prompt mates. Because I'm way more likely to talk to somebody that's part of that community than just like, random on LinkedIn.

Speaker A: Is there anything else that you want to share with the listeners before I let you go?

Speaker B: We're all figuring this out. I don't, I truly don't think that anybody knows what, what's happening in this space, what we're doing. Like, I, I've been fortunate enough that I have. I got exposure to LLMs and MCPs. Two and a half years ago when I rejoined Shopify, I just had this head start, most people didn't, of building and this engineering community of people sharing and Toby the CEO, building things and sharing them in Slack channels and commenting on other people's builds. I just got lucky with the exposure that I was able to get. And I did so many of my builds that were learning and failed builds there. And I'm so thankful for a culture, a team that was so encouraging of experimentation. That's the only reason that I am where, um, I am right now, because I had those opportunities. It's not that I know more or anything like that. I've just been very fortunate and I've taken advantage of those. And so I think that, one, the lesson there is just keep doing it. Everyone's gonna be great if they want to keep in the space. And two, if you're not somewhere that, that you're surrounded by people that are getting to experiment and are, uh, challenging you, like, and I know it's hard in the job market, but, like, try to change that, you know, like, you don't want to be the smartest person in the room. You don't want to be the smartest person on the team. You like this. These are recipes for stagnation and for learning to just stop. And so find ways, even if it's not your job. Maybe it's a community, a, uh, WhatsApp group, whatever. Find ways to make sure that you're serving people that are a lot smarter than you because, one, it's humbling and we all need that. And two, you're just going to learn so much from, from these people. And so, yeah, just like, dive in and learn. It's really, it's fun.

Speaker A: I'm grateful that you, you not only took the time to share Today, but also where we got the chance to originally meet at the AI Recruiting summit that I host every year and coming in and sharing about what's changing. And I remember, uh, a few years ago I was on a run. I was like, you know what? I'm seeing this role as a game changer for recruiters or internal TA and a combination of the agency side and I. It looks like the internal teams took off with it first. But I'm also starting to see the agency world start to play with this like uh, engineer behind the scenes that's putting all these different things together and all these signals. And I'm watching how this is starting to actually move the needle. Like you said earlier, I don't think we're truly there yet across the board. But it's that the opportunity to be that forward mover that, that fast movers, what's these doors. You can literally walk in and start changing your career if you want to be that person. And that's, that's the reason why I wanted to do this episode is I get it. Not every single person wants to sit there and be smiling, dialing. I can four hours a day. What if you could be the person that ends up building the things for your team that actually moves the needle, actually has that roi, actually saves time, actually doubles the amount of revenue, actually doubles the amount of hiring. The opportunity is there and this is the very, very beginning. So Dan, excited that you joined us and for the listeners, this is your year. Choose your path, make it. It's all you. Atlas is a crmx. A CRM that actually understands context. It captures everything you say, hear, read and write. Every interview, client, call, LinkedIn message, email and WhatsApp automatically. Not because you spent the time entering it. It's because Atlas is always listening. Then it goes one step further and tells you the next action to take. So when you need to fill a role, people search, ranks the best candidates and tells you why. No digging required. The same memory turns BD into a short list of who to chase and why. So you win more clients need to see your pipeline, speak to your dashboards. And Atlas builds a view automatically tracking whatever matters to you. Whether you place permanent or contract. Atlas covers both with a new contract suite. That means you never have to chase a timesheet again and live margin visibility on every single placement. And with its new MCP and API, Atlas plugs into your LLMs, Claude, ChatGPT and the rest of your stack. The low value work that keeps you from billing, Atlas does it for you. Atlas customers right now are saying that they're seeing a 50% higher candidate response rate, a 35% increase in new clients, want 15 hours saved every week, monthly billings jumping 85%, and agencies hitting 150% of their annual revenue target by building their business with Atlas. If you want more information, check out Recruit with Atlas.

Speaker B: Thanks for listening to this episode of the Elite Recruiter podcast with Benjamin Mena. If you enjoyed hit, subscribe and leave a rating.