Sourcing School by RecruitingDaily
The team and audience of RecruitingDaily discuss sourcing tools, news, need-to-knows, need-to-use, and a whole lot more. Class is in session.
Sourcing School by RecruitingDaily
Transforming Recruitment: Insights from Emi Labs and RecruitBot
Get ready to revolutionize your approach to hiring! We're bringing you cutting-edge insights from two insightful CEO's - Mateo Cavasoto of Emi Labs and Jeremy Schiff of RecruitBot - who are reshaping the world of recruiting and sourcing.
Mateo will walk you through Emi Labs' pioneering methods to automate communication in the hiring process for frontline workers. Understanding and integrating into a hiring managers' daily routines can make technological adoption smoother and more efficient.
As we pivot our focus to sourcing, Jeremy from RecruitBot reveals how their passive outreach tool, much like Netflix's recommendation system, leverages machine learning and data science to engage with hard-to-fill roles.
We also delve into the importance of customer success for both Emi Labs and RecruitBot. So buckle up for an episode loaded with invaluable tips and insights. Tune in now!
Special mini series recorded with Oleeo at HR Tech 2023 with hosts Ryan Leary, Brian Fink, and Shally Steckerl.
Listen & Subscribe on your favorite platform
Apple | Spotify | Google | Amazon
Visit RecruitingDaily
Twitter @RecruitingDaily
Join the Secret Sourcing Group
Learn more about #HRTX Events
Ladies and gentlemen, boys and girls, party people, welcome back to HR Tech. This is Brian Fink. I'm joined with Shaly Stekrel, the godfather of sourcing. We are rocking the mic here at Olio's podcasting booth live on the floor in Las Vegas. We are joined by not one, but two guests from two different organizations. We've got Matteo from Emmy Labs, right yeah.
Speaker 2:Are we good?
Speaker 1:Awesome sauce. And we've got my man, jeremy, who has already been on the podcast one time, to talk about his sensational sourcing tool for all things digital and environmental, and I love what is her name at Recruit Bot.
Speaker 4:Oh are you referring to Arby.
Speaker 1:Yes, okay, all right, I was, so we're going to talk about who Arby is in a minute, so we'll get into that in just a second. That was what I was struggling for for words, shaly. What's going on, my man? What's the vibe on the floor? It's looking pretty good. Yeah, we're getting, we're getting busy, all right. So, speaking of getting busy, what I want to do is I want to get busy, busy, busy, busy. Be as quick as can be. What I want to ask, I want to ask a use case for the problem that each of you are trying to solve in recruiting and sourcing today who wants to go first? Okay, I'll go first.
Speaker 2:My vote is but damn Okay, I go. Thanks for having me first of all Happy to. So typical use case of ME so we streamline the hiring process for front-line workers and a typical process that we find is that in the recruitment process of hiring of front-line workers, hiring managers are in the field. They're either at a work at a workhouse, at a facility or at a retail store, so getting their availability it's really hard. So we streamlined all the kind of experience using chat. But we found that streamlining the process of the hiring managers it's critical. So typical use case it's the implement ME on top of the HES and they automate communication with the candidates but also internally with hiring managers. So they usually implement ME to reduce the time to hire but specifically to save time from recruiters and hiring managers, impacting revenue, not only productivity and cost All right, so I got it.
Speaker 1:I know I said I was going to do one thing. I'm going to kind of whoop, whoop, whoop and whatever. All right, how do you train managers to use this technology? Because at the end of the day, they're not sitting in their offices surfing LinkedIn to see who's got what status update. How do you?
Speaker 2:train them, first of all, talking to them, understanding their day today. So we work with Walmart, kinecant and on, and we went, talked to them and we understood that they are not in front of a computer. So we asked them which tools do you use? And we found that, for example, they all had phones, text in some populations, like WhatsApp. So we built our tool on top of the things that they were already using and we designed this for them. So if you design from that point of view, then the change management it's always tough. It's a bit easier than trying to push down their throats a new interface, a new system, something that is not designed for them, like a calendar link or something like that. So that's the starting point for us.
Speaker 3:All right, so this is scheduling specifically no, we do end to end.
Speaker 2:From accounting it says hi be a any job board and they are redirected to any within the application in the ATS and then we streamline to retention. But that's the typical use case that we are encountering the hiring manager back and forth, all right.
Speaker 1:I want to jump on that word streamline and I'm going to turn the conversation over to Jeremy to talk about his use case and what. Okay, so real quick. I've already interviewed Jeremy before. I'm fascinated by everything he's done with machine learning. I'm actually sitting across the table from an actual data scientist. So when I talk to people who have built recruiting tools, they're usually coming from a sales side. They're not coming from the technology angle. So tell us about RecruitBot and the streamline that it provides.
Speaker 4:Yeah, always great to be here. I love talking to Recruiting Daily, so really appreciate the opportunity. Again I'm great to be chatting to you guys. So RecruitBot again is a passive outreach tool. So our job is to help engage with very hard to fill roles, specifically in white color recruiting. But that can range from the original problem that I had, which was when I was running machine learning and data science at OpenTable. I could not hire enough machine learning engineers. They're just, they're purple squirrels, they're impossible to hire and the processes that you're usually using.
Speaker 3:Sorry about that. That was me.
Speaker 4:The processes you're usually. Oh yeah, it makes sense Processes.
Speaker 4:You're usually using the processes that you're usually using are pretty manual and it often is requiring really smart sources to run all over the internet and find information from six or seven different sites, to go and cross-reference it and say here's the set of people that I want to go and reach out to.
Speaker 4:And so Recruitbot is ultimately focused on how do I do that all in one spot and how do I make that incredibly easy? So, rather than going to seven different sites to go and find your candidates and reach out to them, you just log into Recruitbot, run a search, then use real machine learning to go and find people who are similar to the people you like. We like to say the same way that Netflix would recommend movies similar to movies you like. You can do that, except we're going to recommend candidates that are similar to the candidates you already like, just giving feedback on a one to five star scale. And because we have all of the contact information, emails and phone numbers, we can automate email campaigns to make it really easy to sort of talk to them. Or you could even load that into your favorite CRM and automate texting or anything else from there. So it makes it really easy to solve the problem suit to nuts of engaging with really hard to fill talent.
Speaker 1:Jeremy, there's one question that I don't know. We've had three conversations. This is our third conversation. I think we're at three now, yeah, oh, all right.
Speaker 3:Hey, we're going to prom. I love it.
Speaker 1:He's keeping count. So one of the things that I Let me keep coming back, One of the things I don't know is that does Recruitmont use natural language search to fund those candidates? If I say hey, I'm looking for data scientists who went to MIT who live in Boston.
Speaker 4:Yep.
Speaker 1:Okay, Maybe I should have made it a little bit harder than MIT in Boston right, yeah. Right.
Speaker 4:So in some ways yes and in some ways no. So we aren't yet literally I can go and type in any generic text string and have it work. We are playing with technologies like that internally, but having that work really robustly for users is really important, and so we'll see sort of how that evolves. But when I refer to the machine learning that's going and predicting which candidates are relevant to you, that is already and has been for many years using machine learning and natural language processing. That's how we're actually going and prioritizing the resume.
Speaker 4:So what's happening here is the system's going and saying I'm trying to generate statistical patterns of what's in common with the candidates that are good and what's in common with the candidates that aren't good. Right, and so this might be obvious, things like oh, these job titles are this amount of years of experience. But it might be much more nuanced. It might be that they have leadership words, they have ownership words or they are very numbers driven if they're a sales role. These are sort of more vague concepts that a natural language processing system can go and understand, and go and understand the patterns at a conceptual level when it's going and working out how to recommend it to you. So we've been integrating that into our system for many, many years now.
Speaker 1:Okay, and one of the things that I think the two of you have in common, even though you're coming at different problems, is the level of customer service and customer success that you're all nodding, that you guys are vested in. What and why is success to the customer important? I know that it's like just signing on the dot of the line, but like, what tangible results are we trying to produce for our customers? Who wants to handle that one first?
Speaker 4:I mean I'm happy to start. I mean tangible results are really easy. They come to us with a very specific problem. They're all these roles. They're really hard to fill. We don't know how to do them. We're not getting in the volume of inbound. We're not getting the right people in through LinkedIn or whatever other mechanisms we're using. What do we do right? Like they're Basically in panic mode. They're like we need to hire these sorts of people. This, this is essential for the business.
Speaker 3:What they just can't find what they're looking for.
Speaker 1:Yeah so it was a reference to you too, with the sphere you can playing at the sphere.
Speaker 3:Yeah, they're playing right now. Yeah, they're playing, yeah, they're gonna be playing tonight.
Speaker 1:I think it's like $8,000 a ticket. Shali, can you front? Yeah? No, you said yep. I heard a yep.
Speaker 3:Yep no.
Speaker 1:Right.
Speaker 3:Yes.
Speaker 1:So, mateo, what about? What about?
Speaker 2:CX, cs, like I think it's critical. I divided, like in two or three Points of contact. The first one is when you implement a platform like CS and being closer to the customer support. It's critical because you need to understand the specific pains. So technology is just like a tool to solve problems. So you need to understand which are the specific pain structures and what they need. And that's a two-way Communication where you teach also maybe the customer which are the best practices, what our customers are doing, and you listen, you try to solve their problems.
Speaker 2:Then you got the change management part, which we discuss a bit when talking about hurry managers, which is technology. Again, it's just one point of what, one starting point. But then you need to make users use an at all, like me, which we're talking about hundreds of recruiters. And then the final piece connect you to the ROI and what we are talking about, which is like Showing the ROI the money and, from a CS point of view, continuing that relationship and like Helping your champions internally to follow up that Financial return, like over and over throughout the entire life cycle of the customer. So for us, like we did many mistakes in the past and we took that and that was a big difference beyond the product.
Speaker 1:All right about big differences. I'm gonna let Charlie get a question in here edge-wise, because I've been taking up the, I've been sucking all the oxygen out of this. What you got, uh. Okay, great, we were talking about the Tell them who you're talking to. You pointed to you. Pointed to Jeremy.
Speaker 3:I pointed Jeremy okay, alright and we were talking about the language. We sort of later what you were talking about before here, that the natural language Recognition and natural language querying. So I just wanted to kind of dive into that a little bit and Find out the the answer to the question that I had asked you before, which is you know, you obviously you're not using, let's call it, off-the-shelf AI type of, you know, large language models. It's not quality I, because it really isn't. You said you've got something different that is, it's a large language model on top of yeah, so we have to be in.
Speaker 4:We haven't launched it yet, so I have to be a little bit cagey about how it's going to work.
Speaker 3:Yeah, but concept conceptually.
Speaker 4:Yeah, I mean effectively it's, it's what we're just talking about earlier, right like the. The way that people want to talk about these things are they just want to say I want to find people that understand this technology and this technology may be a really recently burgeoning technology that we don't have. A lot of people on LinkedIn that have said like hey, this is a skill of mine, right, and that's it's.
Speaker 3:That's exactly the problem, right there is that there's all these keywords that people are used to searching with, and the language changes so quickly that you really would have a very large Boolean statement with all these esoteric terms in it to still only capture a small portion of the population anyways. So that's why I think the benefit of these large language models is be able to expand that ontology, right.
Speaker 4:Totally agree. I mean, they were even talking about it earlier today, about how much they're like. They're like big, like companies are really focused on making sure that they're staying ahead, right. And so if you're staying ahead, you're going and pulling your engineering team and product team and design team and you're saying, hey, what is what's the technology? We need to know.
Speaker 3:We need to know, because that's what we need to hire, for you want to hire.
Speaker 1:I go and then type that keyword into LinkedIn and whatever else and that they're like what do you mean?
Speaker 4:this technology Like I, like I love seeing. Hey, I've been you, I'm a, I'm a generative AI expert. I've been using chat GPT for four years and you're like cat, gpt hasn't been around for four years.
Speaker 3:That's yeah. What are you talking? About and and so, yeah, it's exactly the keywords. Yeah, so so the question that I wanted to get to was the data are you using? Are you looking through homogenous data? Is that's kind of where I not. In homogenous data, things tend to get really messy.
Speaker 4:Yeah, so there are many different steps that are necessary along the way right. So a great example is we have our database of 600 million people.
Speaker 3:But those are people, so that's homogenous.
Speaker 4:Exactly, but that's coming from a number of heterogeneous sources. And then we part of our technology stack is actually going and sort of synthesizing that down into a way that, hey, data source A refers to thinks about data in this way, data source B does it in this way.
Speaker 3:And it's organized.
Speaker 4:And we have to organize and there's a lot of mapping. That's, frankly, just humans in the loop going and cleaning up sort of incredible amounts of data, and then there's having to match this and then there's the being able to do it at scale, right. I think we're dealing with over 200, or two billion data points now to generate our 600 million candidates every six weeks, and we're pulling in more sources literally every time.
Speaker 3:We're building in our data set. Yeah, and that's where large language models have Are going to flourish. Yeah, that's what I'm saying. Yeah, exactly.
Speaker 4:So if you can go and sort of homogenize the data like the way you're framing it, that makes the task for the large language model a lot easier to go and sort of do things on top.
Speaker 3:And predictable.
Speaker 4:That's exactly right. And again, we all have these. Like there's all these problems right where it'll invent.
Speaker 1:Like it'll invent sort of Hallucinations yeah hallucinations.
Speaker 4:A lot of people aren't familiar with the term so I try to avoid it. But yeah, exactly, we got to educate everyone about it. It'll hallucinate all sorts of data that's not there, and so the simpler the problem and it'll sort of, it'll be easier for it to compensate for it.
Speaker 3:The more congruent. That's exactly right.
Speaker 4:And so if you're doing this right, there's going to be a lot of balance between, like there's sort of this naive solution which is I just go into chat GPT and I type in whatever I want and I get my Boolean's ring. That's the misconception and like A, that'll work okay, but to your point before, when we were talking before, it's going to be really general it's going to be super generic and you're not going to get sort of.
Speaker 4:The whole point is that you're narrowing in on what you want. I often talk about the sourcing donut where, like, you don't want to find the center of the donut, where everyone's looking for the MIT software engineer that we're in In Boston.
Speaker 4:Yeah, in Boston that works at Google because, like everyone's already reached out to those people, you want to find the actual donut. You want to find the people who are great for your company but are only okay for sort of everyone else's company, and so the more you like. What we've basically found is we're sort of have a very large advantage because we've spent years building out these incredible data sets and pipelines that allow us to homogenize the data, which then allow us to lay other things like other types of natural language processing technology, on top to sort of provide a lot more power. And if you just threw it at the homogenous data sets, you wouldn't get anything out. Like every now and then, I see someone that's like we have two billion, we have two billion candidate profiles, and I'm like, yep, you did not deduplicate your data. That is what I am hearing. When you have two billion profiles, it is not that you magically have invented candidates that like don't exist in the world.
Speaker 1:You have solved the labor crisis.
Speaker 3:That's right, yeah, you have two billion, but 1.6 billion of those are the same. That is correct. So, carrying it over to Mateo, what patterns are you identifying in your side of?
Speaker 2:the world, so I'll be cautious, because Jeremy here is an expert. So what do I say? No, but since we work with front line workers, we comeropyvs so far Many of these workers is from them. Workers are offline, so there's no data online. Yeah great, so there's no LinkedIn data. Wait a minute, they're not on Facebook. They are on Facebook a lot, but you can spread out under that name and not under like. You don't have detail about their work history, what they want to do, so there's like they are sort of it's Facebook, like every everybody's on Facebook.
Speaker 2:Yeah.
Speaker 1:Unless, I know, I know everybody over 30.
Speaker 2:Yeah, but so for us, like that data said, like we part of what we're doing, it's not getting to the source in space, so we're not looking to get new kind of a right.
Speaker 2:Yeah, but we're looking to to way to explore things on how data enables us to provide value to our customers. So, for example, like this, this workers coming this data from, from from each one of our customers, allows us to understand what's the perfect candidate profile for this company, for this kind of worker. So, for I'm we're talking about our employers. They have like hundreds or thousands of workers, which means that we process like millions of candidates. We got a counter in our booth that it's in real time that shows how many from and candidates will process so far, and it keeps ticking and we are at 7 million candidates right, which making your point. It's not that we can talk generally speaking about data points, but each one of our customers. They got a big chunk of those. So we are playing around to play somehow To see how we can provide unique value to them and what are the?
Speaker 2:yes, so you're finding patterns amongst that, exactly for example and under there are more generic things that we are doing which are more basic, what you are saying, which is, for example, enabling a better Conversational interface with the candidates. So we get to understand with you got, if you got seven thousand, seven million Candidates that went through the similar conversation, you can, in case and understand how that Be more preemptive, exactly, yeah, yeah it's one of the biggest challenges I think with with people in our space and technology and HR it just in general.
Speaker 3:Talent technology is ideation. You know, the technology has existed For a while but people just don't. They haven't been able to figure out how to apply it because they don't really have any ideas on what to apply.
Speaker 2:The problem set to the problem is when you start with that, like what we I agree, complete with you like we thought, first of all we need to resolve, solve a real problem, which is the efficiency of the hiring. Right in the process, we get all this data right, we make these candidates which are invisible, visible, yep. And on top of that, we keep exploring new ways of telling value. But you can start the way around, like getting data and then thinking, okay, what's the value that we'll deliver?
Speaker 3:because at least you can build a company like that first value, you get data and then you explore new ways and it's to stack value on top of that and identifying that essentially figuring out what the problem is that you want to solve is not something that TA people are really good at, because they're looking at the problem head-on, hiring people, but they don't really.
Speaker 1:That's why I ask the questions about what wise the, Not the five wires.
Speaker 3:Yeah yeah, exactly. So you know the problem that people have with LinkedIn nowadays, becoming dependent on it and having nothing else but LinkedIn as their only source. That's not a new problem. Before LinkedIn it was career builder, before that it was monster.
Speaker 3:Before that it was net temps or the fact, or online career center or right, so that there's always been this dependency on that one technology, because it was the ones, or solution, because that was the one solution everybody knew. But the. The origin of the problem isn't where do we go to find candidates, it's when are the candidates?
Speaker 4:And, specifically, how do I actually engage with them right like?
Speaker 3:after you find those exactly like finding like people you want.
Speaker 4:You want a thousand software engineers. Here you go like it's super easy to do. You want. You want a thousand software engineers that'll talk to you very, very different problems, a completely different problem.
Speaker 2:And the real problem is you need the software engineers to run your company, to build product or, in your software, your retail associate to sell on your retail. So you need to go, I think, that deep and then go to the problem like what's the real problem? Okay, having people, then why don't you have the people? Okay, because my Harry Majors tell me that I don't have the people.
Speaker 1:And then you get Because they're not on the All right, all right, we're not gonna beat up on LinkedIn too much tonight. All right, so real quick, just to recap we have been joined by Jeremy at recruit bot. We've been joined by Mateo at Emmy labs. I've been joined by the godfather of sourcing, mr Shally stick. Well, we are coming to you live from the olio booth at HR tech in Las Vegas. We wish you were here. We're sorry that you're not. Thanks for joining us and we'll see you on the next episode.
