October 27, 2023 • 15 mins
Get ready to unravel the future of recruitment with Rich Mendis, the ingenious mind behind HireLogic. We'll uncover how HireLogic's innovative product is redefining recruitment process, engineered to automatically extract crucial interview intel and smoothly integrate it into an ATS. The product's natural language user interface allows you to interact, probe, and juxtapose candidates' responses.
Ever wondered how AI regulations could possibly affect candidate ranking? Rich enlightens us about the prospective evolution and ramifications of this technology. As we navigate the uncharted waters of AI in recruitment, he emphasizes the significance of revisiting this subject in times to come. Prepare for an insightful journey through the transformation of the recruitment landscape.
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
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:07):
Hey everybody, welcome to Recruiting Daily
Sources School podcast.
We are live at HRTech in LasVegas.
It is me, it is Ryan Leary, itis Shaly Steckel and Ryan's
going to introduce our guestfrom HireLogic.
What's going on, Mr Rich Mendez?
Speaker 2 (00:22):
Hey guys, good to be here.
Speaker 1 (00:24):
Rich thanks for coming on we're excited to he
didn't match your energy.
He said hey, Rich, how youdoing?
I'm doing great.
Speaker 2 (00:33):
All right.
Speaker 1 (00:36):
Okay, so Rich, you know it is a the afternoon has
kind of taken a little bit of adip.
There's still an energy,there's still a charisma that's
going on on the floor.
Can you tell the people who didnot make it out to HRTech, who
should have come to HRTech, whatthe vibe is?
What are the feels?
Speaker 2 (00:53):
This is actually my first time and I'm pretty
impressed for an HRTechconference.
We've got a lot of vendors,different sizes from, all the
way from small startups to someof these names which probably
everyone in HR recognizes.
It's nice to see the mix andgood to see all the AI stuff
which we're involved in as well.
Speaker 1 (01:16):
First tick on the bingo card, ai stuff.
We should have done that.
Speaker 2 (01:19):
We should have had a car.
Speaker 1 (01:20):
There you go.
Speaker 2 (01:22):
We're going to do it tomorrow.
Speaker 1 (01:24):
Let's just draw it here.
There's a FedEx in the Luxor.
We can make that happen.
We could make that happen.
All right, the Luxor.
Let's just talk about the Luxor, no no, no, no, let's not do
that.
Speaker 2 (01:33):
Let's talk about making it happen.
Speaker 1 (01:35):
So real quick.
Can you give us a 30,000 footview about what HireLogic does
and why you?
Speaker 2 (01:42):
came out here.
Sure, so we actually we weredoing AI before ChatGPT was
launched, before it was core.
So what we do is we'll listento any live interview, so be it
on Zoom Teams, in person, on thephone, and we will extract all
the intelligence about thatcandidate what are their
(02:03):
potential strengths, concernstheir skills, all those things
as well as information about theinterview.
How much of the job descriptiondid the interview cover?
Did the interviewer ask anyquestions?
They're not supposed to aroundage, gender, race, those types
of things and push that all intoan ATS so that the recruiter or
hiring manager doesn't have togo back in and enter any of that
information.
Speaker 1 (02:23):
So anyway it says, don't have to enter in any of
that information.
Is that information accessiblevia search like a Boolean search
string or?
Speaker 2 (02:32):
even better.
So the search capability is theATS.
It's available, based onhowever the ATS works.
However, what we've done iswe've also put on top of all
this interview intelligence aChatGPT interface, so you can go
up and say, hey, of the sixsales people we interviewed,
which one has the best abilityto prospect on their own, which
(02:53):
is not something you can reallyask in a Boolean search, but
this will interpret that you canask.
Speaker 3 (02:58):
you're just not going to get an answer.
You can ask anything you want.
Speaker 1 (03:02):
Who is the best sales person?
He's being super humble here.
Speaker 3 (03:06):
He tells you how it works, but you've seen it in
action.
Speaker 1 (03:09):
Right, I've seen it in action.
Speaker 3 (03:10):
But like have you seen it?
I don't think so.
You need to see it.
Speaker 1 (03:14):
I do need to see it.
It is sick.
Speaker 3 (03:17):
It's probably the best product software package
I've seen this year.
That's how praise comes from.
We pretty much caught him offin the midstream there, so go
ahead and finish your talk.
No, no.
Speaker 2 (03:28):
I appreciate that.
Yeah, I mean, look, as astartup C-level exec, I've had
to interview a ton of people inmy career and it's self-serving
for me to say this, but this Inever want to do another
interview without it, becauseI'm constantly spending time
after an interview trying togive the candidate their due by
writing the pros and cons andstuff like that, and this does
(03:50):
it for me automatically, which?
Speaker 3 (03:51):
really makes a big difference.
Compare this to thisside-by-side type of thing.
Speaker 2 (03:57):
I think it's a little bit different than that,
because you can ask the samequestions and then compare the
answers and you can do theratings and stuff.
But imagine you've just overthe span of, let's say, three
weeks, you interviewed 20candidates for, say, a customer
service rep.
Now the research shows youremember the first candidate and
the last one, but everyone inbetween is blended together.
(04:17):
So imagine being able to go inand ask questions about the
ideal candidate.
Ask questions about thequestions.
Questions about the questionsor responses, and get that
information in a naturallanguage UI Right.
Speaker 1 (04:32):
Okay, the natural language UI, and you use
customer success as an example,as a jumping off point.
Where do you feel that there'sthe best use case for this
utilization right?
Is it high volume hiring?
Is it front line?
What?
Where do we?
Speaker 2 (04:48):
go here.
So when we launched this we'renot that old.
We've been in the market forless than a year.
Most of the time we spenttraining that machine learning
model to get really good.
We spent about 12 to 18 monthstraining the model, but when we
launched it we felt like itwould be knowledge workers.
That would be the most obviousfit right.
You were.
Quality of hire is important.
(05:09):
You're interviewing, askingthem lots of questions.
Interestingly, we launched apilot with a job board and the
majority of the positions wereall service type workers.
So think of like a PopeyesWendy's Chick-fil-A.
Speaker 1 (05:26):
You're making me hungry right now, like at the
end of the day.
Go ahead.
I'm sorry to interrupt you, myfriend.
Speaker 2 (05:32):
But, yeah, interviewing these people.
And when we talked to them andasked them, hey, what is it
about this, that, why are youusing hire logic?
They said well, the peopledoing these interviews are not
really trained on how to do it,so we like the fact that it can
listen for things they shouldn'task and give them some training
and coaching on.
So it turned out that, yes, theoutput was useful, but the use
(05:55):
case was more for improving thequality of the interview.
Speaker 3 (05:59):
Of the hiring manager itself, who are?
Speaker 2 (06:01):
very well trained in knowledge workers, but not very
well trained in the.
Speaker 3 (06:05):
Supposedly.
Yeah, so you know.
Right, you would think.
Also, I was going to say Idon't know if you found this or
not, but it makes sense to me.
Logically, If you are lookingat candidates that have a very
defined skill set, it's a littlebit easier to, let's just say,
look at homogeneous data andcreate a pattern.
(06:27):
Right, okay, there's 20candidates and they all know how
to write in this particularprogram.
So now you're just looking atthe quality of their programming
and if you can read the codeyou can kind of sort of tell.
But if you've got 20 candidatesand you're interviewing for a
very soft skill like customerservice, there's not like a
(06:47):
keyword that you can look for?
Speaker 2 (06:48):
There really isn't, and so you know it's a lot of
behavioral interviewing.
Hey, have you been in adifficult customer service
situation?
Tell me how you solved it.
Right Now, the response to thatis number one not something
that an interviewer typicallywrites good notes about and then
goes back into an ATS and putsit in.
Speaker 1 (07:05):
And number two Exclamation points.
Awesome a customer service, andit could be any experience.
Any experience.
Speaker 3 (07:10):
And from the experience you can determine if
that was a good customerexperience or not.
Speaker 2 (07:14):
Yeah.
Speaker 3 (07:15):
But there's not a way to like search that Bingo bingo
.
Speaker 2 (07:18):
Yeah, exactly, the other thing we saw in that
particular industry is therecruiters.
So we also work with staffingand recruiting companies, not
just the HR side.
So a large staffing companythat we're talking to has
recruiters that hire forindustrial labor.
So we generate questions forinterviews as well, and one of
the questions they said is hey,put in a CNC machine operator.
(07:42):
One of the other ones wasforklift operator, cause we
don't.
We've never operated forklifts.
What do we ask a forkliftoperator?
Right?
So we generate questions andthe questions are amazing and
they basically covered differenttypes of forklifts and stuff
like that.
So, in addition to asking theright questions, you then get
the analysis which you can mineInteresting.
Speaker 1 (08:03):
What kind of questions?
Okay, so we talked aboutforklift operators and we talked
about white collar.
Speaker 3 (08:08):
Tell me, about a time you drove over someone's foot.
Speaker 1 (08:10):
I'm like I'm sitting here and, like I'm, my mind is
just racing.
I'm like like I know all thequestions that I should be
asking somebody about Python orabout R and how these are
implications, but like your usecase for a forklift operator, it
almost-.
What's the first thing you dowhen you get into the forklift?
No, but it empowers therecruiter to ask and to give a
more personalized experience.
Where we're trying to get towith that, do you offer
(08:34):
recruiter training on how to bea better recruiter?
I'm just wondering from like ananswerly kind of perspective.
Speaker 2 (08:41):
Rich.
We don't offer as a service, wedon't offer that, but we have a
dashboard in our product where,if you're managing a team of
recruiters, you can go intohigher logic and you'll see up
to the last call across yourteam how many interviews were
done, what percentage hadcompliance issues.
You can drill in, see theaverage interview duration and
try and pinpoint.
(09:02):
You know if this is your bestrecruiter.
How are your newer recruitersdoing relative to that?
Speaker 1 (09:06):
first oh, that benchmarking.
That's interesting.
Now where?
Speaker 3 (09:10):
you're seeing compliance issues.
You're actually driving themback to the point of the
compliance issue.
Speaker 2 (09:15):
That's right.
So go back into the interviewand hey, you know why is this
person asking a lot ofinformation about race?
So this actually came up wherewe had an executive recruiting
company and one of the thingsthat one of the recruiters would
like to do is establish arapport with the individual.
So whenever they had aninteresting accent, they would
say oh, that's an interestingaccent, where are you from?
Originally Innocent enough, butnot something you should ask.
(09:38):
And our software picked that upas a potential and we'll bring
them back to that point in theconversation, correct?
Speaker 1 (09:47):
Used for training.
Oh, I like to use case fortraining I love how this could
be used.
Like you know, I think about.
I'm selfish.
I immediately think about myteam and how we can up level my
organization, right, you know,that's where I think about that
first, but then I think aboutthe broader implication as to
how this affects RPOs.
(10:07):
Right, because RPOs are sent onan assignment to recruit and
spin up very quickly.
You need them to be excellentat what they do and you need
very little training time.
But this offers that elementthat says this is how they're
using the system, this is whythey're using the system, these
are the questions that they'reproviding.
It gets me really excited aboutthe RPO space, and I don't
(10:28):
usually get excited about RPOs.
Nobody does no.
Speaker 3 (10:32):
Ryan, just because you had a bad experience, my
experience is done.
Speaker 1 (10:37):
My experience is done and dusted, all right.
Speaker 3 (10:39):
It's still seven, two going into the seventh.
Speaker 1 (10:42):
My phone says that it's eight to two going into the
seventh.
Oh, mine does say eight to twonow.
Okay, so just real quick ifyou're tuning in Three weeks
later.
This is Ryan and Brian watchingthe Phillies and Brains game as
we're having this conversationwith Rich.
So it doesn't look like theBraves are gonna pull through,
unless there are.
So okay, so moving on.
(11:04):
If you come back tonight, youdeserve to take the series.
I appreciate that.
So we've talked a little bitabout generative AI.
We've talked about trends intraining.
What trends do you see on thehorizon for 2024?
Speaker 2 (11:20):
So, first of all, I think what everyone's gonna have
to get ready for is AIregulation that's coming.
Oh wow, you're the first guestwho said that all day.
That's right.
I think people are not.
They're not paying enoughattention to it and it's gonna
come.
And it's gonna come pretty hardin HR because all of the
regulations that are beingdiscussed in AI are around
discrimination and privacy andthat's right in the heart of HR.
Speaker 3 (11:41):
But yeah, there's already lawsuits Already
lawsuits, right people.
Speaker 2 (11:45):
We won't name names, but have already been hit here
where you can't use AI for videointerviews and stuff like that
anymore.
But what's interesting is, Ithink there's an opportunity for
HR leaders to actually step upin the organizations and be sort
of the people who set thegovernance and policy for how HR
, how AI, is used, not just inHR, but across the company.
Speaker 1 (12:06):
Okay, so across the company.
What about these state and cityand municipal?
Well, state and municipal, andthere's the problem, yeah, is
that an organization can only doso much?
And there's the problem, yeah.
Speaker 2 (12:21):
And HR's been dealing with these things for years,
right or not, these are, orostensibly not, where you're not
supposed to ask these questionsin this state, or you you know
this is the holiday laws in thisparticular state, right?
It's?
That's why I guess we have someof our ADP paychecks and other
companies to help with thosethings.
Speaker 3 (12:39):
Excellent, yeah, but basically it's a bias question,
right?
Yeah, it's a matter of bias andany way that you introduce bias
into the process.
Speaker 2 (12:47):
That's right.
Speaker 3 (12:48):
You're going to have some kind of repercussion, and
in this case it's uncontrolledbias.
Speaker 2 (12:54):
That's right.
Because you don't know what thebias is, and so you know one of
the laws are being very clearthat you can't have AI tools
make the decision on behalf ofthe human.
So even when we present data,we're very careful about how we
label it.
It's a potential concern, asopposed to giving a score or a
ranking or something thatdetermines whether a candidate
(13:15):
should proceed to the next level.
Speaker 3 (13:17):
Do you think Because there's the hidden bias?
Speaker 2 (13:19):
Yeah, so now, rich, do you think that the companies
that are ranking candidatestoday is that going to have to
go away?
They're already in violation ofthe New York City law 144,.
Yeah, yep, so basicallyLinkedIn, I had the LinkedIn
rank.
Speaker 3 (13:34):
Yeah, it's results one through 10 and then 2330.
That's a rank.
Yeah, if they have some, yeah,just think about it.
Speaker 2 (13:41):
Yeah, that's true.
I mean, if they're using AI todo matching and ranking, then
yeah, that could be subject tothat one.
It's interesting.
Speaker 1 (13:48):
I don't know.
I kind of think about some ofthe other tools that I've seen
about candidate ranking thatranks a candidate as an A
candidate or a B candidate or aC candidate, and then
undetermined variable, whichthrows me for a curve ball
because I'm like why don't youjust rank them as a D or an F?
Speaker 3 (14:05):
But at what point is this over-regulated?
Speaker 2 (14:07):
So great question Because you have to be able to
do so.
Here's the problem with thisregulation, though.
Right Is all the research showsthat humans are extremely
biased, whether it's scanningresumes, ranking candidates,
what have you, it's heavilybased on Godin state.
So now you've got this AI toolthat ostensibly maybe improves
(14:27):
it a little bit Learn for humans, but you're putting all this
regulation on it, so are youactually killing a solution that
might help with the problembefore it even has a chance to
mature?
Speaker 1 (14:39):
All right, that's a good question.
That's a question that I thinkthat Rich means that we should
have you back on the sourcingschool podcast in the next six
months to kind of see wherethings are headed and where
things have developed.
I'm Brian Fink, this is RyanLeary, that is Shelley Stackroll
and this is Rich from HireLogic.
Thanks for joining us on Olio'spresentation of the sourcing
(15:00):
school podcast.
I'll see you in a minute.
Hey everybody, welcome to Recruiting Daily
Sources School podcast.
We are live at HRTech in LasVegas.
It is me, it is Ryan Leary, itis Shaly Steckel and Ryan's
going to introduce our guestfrom HireLogic.
What's going on, Mr Rich Mendez?
Speaker 2 (00:22):
Hey guys, good to be here.
Speaker 1 (00:24):
Rich thanks for coming on we're excited to he
didn't match your energy.
He said hey, Rich, how youdoing?
I'm doing great.
Speaker 2 (00:33):
All right.
Speaker 1 (00:36):
Okay, so Rich, you know it is a the afternoon has
kind of taken a little bit of adip.
There's still an energy,there's still a charisma that's
going on on the floor.
Can you tell the people who didnot make it out to HRTech, who
should have come to HRTech, whatthe vibe is?
What are the feels?
Speaker 2 (00:53):
This is actually my first time and I'm pretty
impressed for an HRTechconference.
We've got a lot of vendors,different sizes from, all the
way from small startups to someof these names which probably
everyone in HR recognizes.
It's nice to see the mix andgood to see all the AI stuff
which we're involved in as well.
Speaker 1 (01:16):
First tick on the bingo card, ai stuff.
We should have done that.
Speaker 2 (01:19):
We should have had a car.
Speaker 1 (01:20):
There you go.
Speaker 2 (01:22):
We're going to do it tomorrow.
Speaker 1 (01:24):
Let's just draw it here.
There's a FedEx in the Luxor.
We can make that happen.
We could make that happen.
All right, the Luxor.
Let's just talk about the Luxor, no no, no, no, let's not do
that.
Speaker 2 (01:33):
Let's talk about making it happen.
Speaker 1 (01:35):
So real quick.
Can you give us a 30,000 footview about what HireLogic does
and why you?
Speaker 2 (01:42):
came out here.
Sure, so we actually we weredoing AI before ChatGPT was
launched, before it was core.
So what we do is we'll listento any live interview, so be it
on Zoom Teams, in person, on thephone, and we will extract all
the intelligence about thatcandidate what are their
(02:03):
potential strengths, concernstheir skills, all those things
as well as information about theinterview.
How much of the job descriptiondid the interview cover?
Did the interviewer ask anyquestions?
They're not supposed to aroundage, gender, race, those types
of things and push that all intoan ATS so that the recruiter or
hiring manager doesn't have togo back in and enter any of that
information.
Speaker 1 (02:23):
So anyway it says, don't have to enter in any of
that information.
Is that information accessiblevia search like a Boolean search
string or?
Speaker 2 (02:32):
even better.
So the search capability is theATS.
It's available, based onhowever the ATS works.
However, what we've done iswe've also put on top of all
this interview intelligence aChatGPT interface, so you can go
up and say, hey, of the sixsales people we interviewed,
which one has the best abilityto prospect on their own, which
(02:53):
is not something you can reallyask in a Boolean search, but
this will interpret that you canask.
Speaker 3 (02:58):
you're just not going to get an answer.
You can ask anything you want.
Speaker 1 (03:02):
Who is the best sales person?
He's being super humble here.
Speaker 3 (03:06):
He tells you how it works, but you've seen it in
action.
Speaker 1 (03:09):
Right, I've seen it in action.
Speaker 3 (03:10):
But like have you seen it?
I don't think so.
You need to see it.
Speaker 1 (03:14):
I do need to see it.
It is sick.
Speaker 3 (03:17):
It's probably the best product software package
I've seen this year.
That's how praise comes from.
We pretty much caught him offin the midstream there, so go
ahead and finish your talk.
No, no.
Speaker 2 (03:28):
I appreciate that.
Yeah, I mean, look, as astartup C-level exec, I've had
to interview a ton of people inmy career and it's self-serving
for me to say this, but this Inever want to do another
interview without it, becauseI'm constantly spending time
after an interview trying togive the candidate their due by
writing the pros and cons andstuff like that, and this does
(03:50):
it for me automatically, which?
Speaker 3 (03:51):
really makes a big difference.
Compare this to thisside-by-side type of thing.
Speaker 2 (03:57):
I think it's a little bit different than that,
because you can ask the samequestions and then compare the
answers and you can do theratings and stuff.
But imagine you've just overthe span of, let's say, three
weeks, you interviewed 20candidates for, say, a customer
service rep.
Now the research shows youremember the first candidate and
the last one, but everyone inbetween is blended together.
(04:17):
So imagine being able to go inand ask questions about the
ideal candidate.
Ask questions about thequestions.
Questions about the questionsor responses, and get that
information in a naturallanguage UI Right.
Speaker 1 (04:32):
Okay, the natural language UI, and you use
customer success as an example,as a jumping off point.
Where do you feel that there'sthe best use case for this
utilization right?
Is it high volume hiring?
Is it front line?
What?
Where do we?
Speaker 2 (04:48):
go here.
So when we launched this we'renot that old.
We've been in the market forless than a year.
Most of the time we spenttraining that machine learning
model to get really good.
We spent about 12 to 18 monthstraining the model, but when we
launched it we felt like itwould be knowledge workers.
That would be the most obviousfit right.
You were.
Quality of hire is important.
(05:09):
You're interviewing, askingthem lots of questions.
Interestingly, we launched apilot with a job board and the
majority of the positions wereall service type workers.
So think of like a PopeyesWendy's Chick-fil-A.
Speaker 1 (05:26):
You're making me hungry right now, like at the
end of the day.
Go ahead.
I'm sorry to interrupt you, myfriend.
Speaker 2 (05:32):
But, yeah, interviewing these people.
And when we talked to them andasked them, hey, what is it
about this, that, why are youusing hire logic?
They said well, the peopledoing these interviews are not
really trained on how to do it,so we like the fact that it can
listen for things they shouldn'task and give them some training
and coaching on.
So it turned out that, yes, theoutput was useful, but the use
(05:55):
case was more for improving thequality of the interview.
Speaker 3 (05:59):
Of the hiring manager itself, who are?
Speaker 2 (06:01):
very well trained in knowledge workers, but not very
well trained in the.
Speaker 3 (06:05):
Supposedly.
Yeah, so you know.
Right, you would think.
Also, I was going to say Idon't know if you found this or
not, but it makes sense to me.
Logically, If you are lookingat candidates that have a very
defined skill set, it's a littlebit easier to, let's just say,
look at homogeneous data andcreate a pattern.
(06:27):
Right, okay, there's 20candidates and they all know how
to write in this particularprogram.
So now you're just looking atthe quality of their programming
and if you can read the codeyou can kind of sort of tell.
But if you've got 20 candidatesand you're interviewing for a
very soft skill like customerservice, there's not like a
(06:47):
keyword that you can look for?
Speaker 2 (06:48):
There really isn't, and so you know it's a lot of
behavioral interviewing.
Hey, have you been in adifficult customer service
situation?
Tell me how you solved it.
Right Now, the response to thatis number one not something
that an interviewer typicallywrites good notes about and then
goes back into an ATS and putsit in.
Speaker 1 (07:05):
And number two Exclamation points.
Awesome a customer service, andit could be any experience.
Any experience.
Speaker 3 (07:10):
And from the experience you can determine if
that was a good customerexperience or not.
Speaker 2 (07:14):
Yeah.
Speaker 3 (07:15):
But there's not a way to like search that Bingo bingo
.
Speaker 2 (07:18):
Yeah, exactly, the other thing we saw in that
particular industry is therecruiters.
So we also work with staffingand recruiting companies, not
just the HR side.
So a large staffing companythat we're talking to has
recruiters that hire forindustrial labor.
So we generate questions forinterviews as well, and one of
the questions they said is hey,put in a CNC machine operator.
(07:42):
One of the other ones wasforklift operator, cause we
don't.
We've never operated forklifts.
What do we ask a forkliftoperator?
Right?
So we generate questions andthe questions are amazing and
they basically covered differenttypes of forklifts and stuff
like that.
So, in addition to asking theright questions, you then get
the analysis which you can mineInteresting.
Speaker 1 (08:03):
What kind of questions?
Okay, so we talked aboutforklift operators and we talked
about white collar.
Speaker 3 (08:08):
Tell me, about a time you drove over someone's foot.
Speaker 1 (08:10):
I'm like I'm sitting here and, like I'm, my mind is
just racing.
I'm like like I know all thequestions that I should be
asking somebody about Python orabout R and how these are
implications, but like your usecase for a forklift operator, it
almost-.
What's the first thing you dowhen you get into the forklift?
No, but it empowers therecruiter to ask and to give a
more personalized experience.
Where we're trying to get towith that, do you offer
(08:34):
recruiter training on how to bea better recruiter?
I'm just wondering from like ananswerly kind of perspective.
Speaker 2 (08:41):
Rich.
We don't offer as a service, wedon't offer that, but we have a
dashboard in our product where,if you're managing a team of
recruiters, you can go intohigher logic and you'll see up
to the last call across yourteam how many interviews were
done, what percentage hadcompliance issues.
You can drill in, see theaverage interview duration and
try and pinpoint.
(09:02):
You know if this is your bestrecruiter.
How are your newer recruitersdoing relative to that?
Speaker 1 (09:06):
first oh, that benchmarking.
That's interesting.
Now where?
Speaker 3 (09:10):
you're seeing compliance issues.
You're actually driving themback to the point of the
compliance issue.
Speaker 2 (09:15):
That's right.
So go back into the interviewand hey, you know why is this
person asking a lot ofinformation about race?
So this actually came up wherewe had an executive recruiting
company and one of the thingsthat one of the recruiters would
like to do is establish arapport with the individual.
So whenever they had aninteresting accent, they would
say oh, that's an interestingaccent, where are you from?
Originally Innocent enough, butnot something you should ask.
(09:38):
And our software picked that upas a potential and we'll bring
them back to that point in theconversation, correct?
Speaker 1 (09:47):
Used for training.
Oh, I like to use case fortraining I love how this could
be used.
Like you know, I think about.
I'm selfish.
I immediately think about myteam and how we can up level my
organization, right, you know,that's where I think about that
first, but then I think aboutthe broader implication as to
how this affects RPOs.
(10:07):
Right, because RPOs are sent onan assignment to recruit and
spin up very quickly.
You need them to be excellentat what they do and you need
very little training time.
But this offers that elementthat says this is how they're
using the system, this is whythey're using the system, these
are the questions that they'reproviding.
It gets me really excited aboutthe RPO space, and I don't
(10:28):
usually get excited about RPOs.
Nobody does no.
Speaker 3 (10:32):
Ryan, just because you had a bad experience, my
experience is done.
Speaker 1 (10:37):
My experience is done and dusted, all right.
Speaker 3 (10:39):
It's still seven, two going into the seventh.
Speaker 1 (10:42):
My phone says that it's eight to two going into the
seventh.
Oh, mine does say eight to twonow.
Okay, so just real quick ifyou're tuning in Three weeks
later.
This is Ryan and Brian watchingthe Phillies and Brains game as
we're having this conversationwith Rich.
So it doesn't look like theBraves are gonna pull through,
unless there are.
So okay, so moving on.
(11:04):
If you come back tonight, youdeserve to take the series.
I appreciate that.
So we've talked a little bitabout generative AI.
We've talked about trends intraining.
What trends do you see on thehorizon for 2024?
Speaker 2 (11:20):
So, first of all, I think what everyone's gonna have
to get ready for is AIregulation that's coming.
Oh wow, you're the first guestwho said that all day.
That's right.
I think people are not.
They're not paying enoughattention to it and it's gonna
come.
And it's gonna come pretty hardin HR because all of the
regulations that are beingdiscussed in AI are around
discrimination and privacy andthat's right in the heart of HR.
Speaker 3 (11:41):
But yeah, there's already lawsuits Already
lawsuits, right people.
Speaker 2 (11:45):
We won't name names, but have already been hit here
where you can't use AI for videointerviews and stuff like that
anymore.
But what's interesting is, Ithink there's an opportunity for
HR leaders to actually step upin the organizations and be sort
of the people who set thegovernance and policy for how HR
, how AI, is used, not just inHR, but across the company.
Speaker 1 (12:06):
Okay, so across the company.
What about these state and cityand municipal?
Well, state and municipal, andthere's the problem, yeah, is
that an organization can only doso much?
And there's the problem, yeah.
Speaker 2 (12:21):
And HR's been dealing with these things for years,
right or not, these are, orostensibly not, where you're not
supposed to ask these questionsin this state, or you you know
this is the holiday laws in thisparticular state, right?
It's?
That's why I guess we have someof our ADP paychecks and other
companies to help with thosethings.
Speaker 3 (12:39):
Excellent, yeah, but basically it's a bias question,
right?
Yeah, it's a matter of bias andany way that you introduce bias
into the process.
Speaker 2 (12:47):
That's right.
Speaker 3 (12:48):
You're going to have some kind of repercussion, and
in this case it's uncontrolledbias.
Speaker 2 (12:54):
That's right.
Because you don't know what thebias is, and so you know one of
the laws are being very clearthat you can't have AI tools
make the decision on behalf ofthe human.
So even when we present data,we're very careful about how we
label it.
It's a potential concern, asopposed to giving a score or a
ranking or something thatdetermines whether a candidate
(13:15):
should proceed to the next level.
Speaker 3 (13:17):
Do you think Because there's the hidden bias?
Speaker 2 (13:19):
Yeah, so now, rich, do you think that the companies
that are ranking candidatestoday is that going to have to
go away?
They're already in violation ofthe New York City law 144,.
Yeah, yep, so basicallyLinkedIn, I had the LinkedIn
rank.
Speaker 3 (13:34):
Yeah, it's results one through 10 and then 2330.
That's a rank.
Yeah, if they have some, yeah,just think about it.
Speaker 2 (13:41):
Yeah, that's true.
I mean, if they're using AI todo matching and ranking, then
yeah, that could be subject tothat one.
It's interesting.
Speaker 1 (13:48):
I don't know.
I kind of think about some ofthe other tools that I've seen
about candidate ranking thatranks a candidate as an A
candidate or a B candidate or aC candidate, and then
undetermined variable, whichthrows me for a curve ball
because I'm like why don't youjust rank them as a D or an F?
Speaker 3 (14:05):
But at what point is this over-regulated?
Speaker 2 (14:07):
So great question Because you have to be able to
do so.
Here's the problem with thisregulation, though.
Right Is all the research showsthat humans are extremely
biased, whether it's scanningresumes, ranking candidates,
what have you, it's heavilybased on Godin state.
So now you've got this AI toolthat ostensibly maybe improves
(14:27):
it a little bit Learn for humans, but you're putting all this
regulation on it, so are youactually killing a solution that
might help with the problembefore it even has a chance to
mature?
Speaker 1 (14:39):
All right, that's a good question.
That's a question that I thinkthat Rich means that we should
have you back on the sourcingschool podcast in the next six
months to kind of see wherethings are headed and where
things have developed.
I'm Brian Fink, this is RyanLeary, that is Shelley Stackroll
and this is Rich from HireLogic.
Thanks for joining us on Olio'spresentation of the sourcing
(15:00):
school podcast.
I'll see you in a minute.
Popular Podcasts
On Purpose with Jay Shetty
I’m Jay Shetty host of On Purpose the worlds #1 Mental Health podcast and I’m so grateful you found us. I started this podcast 5 years ago to invite you into conversations and workshops that are designed to help make you happier, healthier and more healed. I believe that when you (yes you) feel seen, heard and understood you’re able to deal with relationship struggles, work challenges and life’s ups and downs with more ease and grace. I interview experts, celebrities, thought leaders and athletes so that we can grow our mindset, build better habits and uncover a side of them we’ve never seen before. New episodes every Monday and Friday. Your support means the world to me and I don’t take it for granted — click the follow button and leave a review to help us spread the love with On Purpose. I can’t wait for you to listen to your first or 500th episode!
The Breakfast Club
The World's Most Dangerous Morning Show, The Breakfast Club, With DJ Envy And Charlamagne Tha God!
The Joe Rogan Experience
The official podcast of comedian Joe Rogan.