Sourcing School by RecruitingDaily

Harnessing AI for Effective Talent Sourcing with Alper Tekin of Findem

October 27, 2023 Brian Fink, Ryan Leary, and Shally Steckerl
Sourcing School by RecruitingDaily
Harnessing AI for Effective Talent Sourcing with Alper Tekin of Findem
Show Notes Transcript Chapter Markers

Are you ready for a transformative look at how technology is reshaping the world of recruiting? Our guest Alper Tekin, Chief Product Officer at Findem, takes us through the importance of data in recruiting, ATS rediscovery, and how his platform is mapping an impressive 750 million people data against 4 million companies to reveal the true picture of a candidate's potential.

We then switch gears to examine how artificial intelligence is revolutionizing candidate search, with Fundum's platform taking center stage. Learn how it automates the sourcing process, all while considering specifics such as diversity data, experience, and project-level skills.

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

Speaker 1:

Welcome back party people. We're coming to. You live from the talent acquisition content lounge powered by Olio. It's recruiting daily source in school. It's me, Brian Fink, it's him Ryan Leary. What's up Right? What's going on? Not?

Speaker 2:

much man. We've got a what do you call it a concert?

Speaker 1:

maybe We've got a woman who is talking. She's trying to hype everybody up to believing that Taylor Swift is coming and just to let you know, taylor Swift is not at HR Tech.

Speaker 2:

She's not here. You two is here. You two is here. Yeah, and they're playing at the Sphere.

Speaker 1:

That is awesome. Yeah, speaking of people who are playing, we are playing the 20 questions game with Alper, who is the Chief Product Officer at Fundum. What's going on, alper? How are?

Speaker 2:

you All right, all right. How's the intro, are you?

Speaker 3:

like it? Yeah, geez okay.

Speaker 1:

All right, 20 questions. Are you ready?

Speaker 2:

No, that was question number one. Question number two Is a hot dog, a sandwich.

Speaker 3:

Probably, probably. Oh, he didn't even think about it.

Speaker 1:

All right.

Speaker 2:

He's thought about that.

Speaker 1:

Yeah, he's thought about that. What's another good?

Speaker 2:

one Toilet paper over or under, so hold on.

Speaker 1:

Jesus, it's like a mullet or a beard right. That's what. That's what I try to teach Maddie.

Speaker 3:

All right.

Speaker 1:

so, alper, it's a beard. If it goes over, if it goes under, it's a mullet. Yes, okay, that is what I'm teaching my seven-year-old Alper's looking at me going. I'm not letting you anywhere near my side right, that's awesome. Alper. You know our last guest that was on. We were talking about the sourcing crisis that's taking place right now, because a lot of recruiters are having to source for passive talent as opposed to having a sorcerer to support them. I immediately think I find them that, you guys. I mean that's what you do. You turn recruiters, you give them superpowers, your little robot that sits on their shoulder. What are you seeing coming down the pipeline for recruiters and sorcerers? What's going on there?

Speaker 3:

First of all, let me acknowledge that I'm with experts here. We're coming from the technology side. In the last three years or so, we've been trying to surround ourselves with domain experts just to understand these pain points more. Once you understand the pain point, like we had the technology down so we can, like you know, create those workflows et cetera. And, as you probably know and I'm happy to go into detail we have this 3D data platform that we've been empowering since the inception of the company, which is like the core differentiator. So, basically, for the, what we've been doing in the last year was, like we're more known as the sort of outbound company. You know, just if you in the hot market, obviously you need that and that's sold quite well. But we also realized that, you know, as things slowed down, like you know, 80% of hiring started to become from inbound sources. So we actually like started leveraging the platform to inbound and outbound workflows. This is, like you know, rate all the candidates I'm like score candidates, scoring ATS, refresh and Rediscovery. We built our own CRM, all those things.

Speaker 1:

Why is ATS Rediscovery important? It just for the people who think that they need to go and find a new candidate every time, jump into this yeah.

Speaker 3:

I mean again, like they already show the affinity to your company, they will listen when you sort of hit them up. They will have higher likelihood of, like you know, opening that email, opening that LinkedIn message, et cetera. And it's much cheaper. I mean to be honest, like you already have that data, you don't have to go like advertise and all that sort of stuff is just like nonsense if you have the data.

Speaker 2:

It's faster, it's cheaper.

Speaker 3:

Faster, cheaper, opens better and probably you know someone that's interested in you and not like someone you're chasing. So there's also psychology attached to it and the issue is that you know especially large companies. They've been emessing, like you know. I don't know like Microsoft existed for like 30 plus 40 years, like, imagine, like a data asset they have.

Speaker 1:

Oh, they've got to have a database for at least a million people.

Speaker 3:

Per month. I would say Everyone applies right. So like, how do you actually? There is a bunch of issues there. First it gets stale right. So maybe I applied to I don't know a company like 15 years ago, 10 years ago.

Speaker 1:

A lot of things change.

Speaker 3:

Even like three months ago, things change. Number one, number two ATSs don't necessarily have good search capabilities to the degree of finding right. Tell me more, Okay, so let's now Keyword searching.

Speaker 1:

Keyword searching. I'm not going to go down that rabbit hole today. Go ahead. I'm sorry.

Speaker 3:

There's nothing wrong with keyword Like the concept we're trying to introduce and hopefully doing an okay job. Is this 3D data right? So it's not just about what I tell about myself or what I don't tell about myself, but do have. It's more like what's the objective truth out there. So the way to get that in our model is like we have maps 750 million people data against 4 million company data over a timeline. So why is that interesting? So I may say I'm a startup person, I love startups. I love startups on my LinkedIn resume, but I may not have been on a startup Like and what is a startup? Right, it's like a LinkedIn startup. It's like an OIL startup. It's find them a startup. What you can do is actually there's an objective truth to that. Like if a company has received series A, series B, series C, series like a D, like those are objective states of a company. So what you can do if you map those data, you can say, hey, this person existed at this company as a sales person between series A's and series C, as opposed to this person as a startup person Opposite example. I found a couple companies. You know I still have to work, so like nothing super successful.

Speaker 1:

But like.

Speaker 3:

I won't say like I'm a startup model or anything like that on my resume, like I started companies, et cetera. There's no keyword that says startup, startup, startup. They would differentiate me from someone else, but actually I find them. System will know this guy started companies or existed in companies when they were series A and scale them. So that's one example of, like you know, objectifying, like the data out there, so you can run things like give me a list of Python developers who understand the chaos of a startup, ie, like they exist in a startup, a startup.

Speaker 1:

B series A.

Speaker 3:

Same thing with Python or Java developers. Right, I may say Python, python, python, java, java. Is it the coffee that I'm talking about? Is it like the snake that I'm talking about or is it like? So? What we do in that case is like we map you to your GitHub profile or, like you know, your Stack Overflow profile and say, okay, this person has actually committed like a Python or Java code. So we know objectively, this person has done this. So the source doesn't have to have that paranoia of going back and forth. Does he really have startup? Does he really have like the verified code skills and that sort of generalizes to like 10,000 or so attributes? No-transcript. Is this an athlete? Does this person have a blog? Does this person have like a? Did this person finish their PhD quickly, you know? Are they? Do they have publications in like molecular biology? Do they? Are they sighted? Have they built diverse teams? Like this is all data that's out there.

Speaker 1:

This is all within the final.

Speaker 2:

Yes, well, the data is out there. Well, the data is out there, but they're making sense of it. Yeah, yeah, yeah.

Speaker 1:

I want to dial down on a dichotomy that you created. You did the specificity in the topography for engineers and you also mentioned salespeople. Those are two diametrically opposed groups of people. How does FINDOM know the difference? Or can you spill that secret sauce?

Speaker 3:

Can you tip the tea Without sounding too controversial, like we don't necessarily go after a keyword when we scrape the public data, like there is a layer that collects the data and then you derive meaning from it. For instance, I don't say like go get me Enterprise Account Execs. I say, get me all titles that you can publicly find in the world, and then you may come and be interested in account executives, you may be interested in developers, someone else may be interested in HR people, etc. So we have curated hundreds and thousands of titles, obviously, and all that sort of stuff. And in the case of Enterprise Account Execs, for instance, you can look up things like have they been a president's club person? Have they been groomed under a really good CRO? Have they worked at, say, a few startups, then a big company and then a startup? Like you can set all the patterns like a CRO would hire and then express those Wow, express those in, actual, our GenAI interface. Like you can just say what you want.

Speaker 1:

Sure.

Speaker 3:

Literally say or speak.

Speaker 1:

And then we'll generate MIT engineers in Boston who went to, who have experience in data science, that weighted tables at a pizza restaurant.

Speaker 3:

The last bit is a bit tricky.

Speaker 1:

Okay, all right, I just, I, just, I just see how far we can go you got 85% of the data, though.

Speaker 2:

Right. So you're the. You know the. Something you said, though, albert. What was really interesting to me is when, so we're able to search for a manager, say manager-level individual, who you know, based on information and or charts. In the company, they've built diverse teams, so we can say I want a middle manager, software developer, whatever it's going to be, who has experience, who has built a diverse team, and then you can, you can pull that information in and make that. I can break it down. I think you like that, yeah.

Speaker 3:

I mean just to demonstrate the point. If you enter an organization and then the the breakdown of the organization, we have diversity data as well. Well, let's say 50-50. And like you know, you're in the tenure for another like three years and it's like now, you know, I don't know like a more equal, like a better sort of team. You know that data, the data is out there but, like it's very, very hard to mine that data as an individual Because, like you know, as a source you have like 15, 20, and 200, 200, 300 candidates each. The complexity is huge, right, so you need the automation there to serve you up and the hiring manager, like, will tell you like one line like hey, I need like B2B salespeople. Like you know the experience. you work for this guy and this guy was the CRO there in Seattle. That's it. It's like a sentence or two.

Speaker 1:

Yeah.

Speaker 3:

You express that and the the source can just see like the candidate that exactly meet those criteria.

Speaker 2:

Now can you get down to project level, meaning a developer at a company worked on this particular specific project at that competitor?

Speaker 3:

If it's publicly available, you will have that against the candidate. And what we recently have done is the following Like let's get into AI a little bit. Let me actually like follow up on the ATS. Why is that important conversation? If you have this data and we refresh your ATS data, which we do, meaning like we look at everyone we refresh them, like make them current and make it searchable, now you can search with this kind of depth that we're talking about not just in the outbound. You can search for these candidates in your inbound applicants as they come in and say, hey, this is a good fit, this is a great fit, this is not a fit In your ATS. You can sort of search the entire, refresh the ATS and, like, bring candidates to the front.

Speaker 1:

Yeah.

Speaker 3:

In your CRM anyone that was in our talent community that fits the bill in your alumni. Meaning did we ever have people that fit the bill that used to work for us in your own company? Like, do we have people that fit the bill in my company? As we speak, All that gets served up in one flow. That's the beautiful thing about the consolidation that the platform enables you, Right.

Speaker 1:

Interesting. That is food for thought. All right, so we're kind of wrapping up on time here. You lead product and find them. Can you give us an idea of what's going to? I'm asking you to spill the tea you know, I know and we're, you know. What's happening in the next six months? What's the roadmap?

Speaker 3:

Yeah, big surprise coming right. Ok, big surprise For which UK? Yeah, ok, I'll remiss when I say yeah, I don't.

Speaker 1:

Yeah, ok if you can't. You can't, and I understand that. No, no, no, but you've got to promise me that you'll come back on and you'll do the show again with Ryan and I when it's time to make that announcement.

Speaker 3:

Yeah, we'll make. We're getting into workflow and how to think about workflows, like more superpowers for the sources, basically Because we sort of mapped out 25 flows that they do day to day and parts of that we can just automate, so they don't have to do this data piping, just background engineering type of stuff.

Speaker 1:

Oh OK, cool, I got you.

Speaker 3:

So that stuff can be automated. And then what we're trying to do is really, really lessen the load on them so they can really focus on the candidate experience, and the way we do this is through AI-powered interfaces, et cetera.

Speaker 1:

OK, we'll talk more again. There's so much for having Six months. We'll have you on the big show. All right, it's Ryan. It's Ryan. We are the talent acquisition content lounge that is powered by Olio, with Recruiting Daily, sourcing School, and it has been my pleasure to have the Chief Product Officer from Findem, mr Alper. Thanks for joining us, sir.

Recruiting Tech, Data, and Sourcing Strategies
Using AI to Enhance Candidate Search
Product Roadmap and Workflow Automation