Introduction

Hi, everyone. My name is Peter, and today I'd like to talk to you about hiring -related challenges.

Do you hear me at the back, by the way? Okay, perfect.

Hiring Challenges in the Era of AI

Hiring -related challenges in the era of AI. I think we can all agree on that there is absolutely no way back.

1Everyone is using AI, both on the candidate side and also on the hiring side, which creates, obviously, some friction on both sides.

maybe I can also start with a poll like hands up who ever had a terrible hiring experience as a job candidate in their life okay what I can tell you the struggle is real on the job candy on the on the candidate side but it's equally bad on the hiring side as well and I brought you a number that that kind of

indicates that the struggle is real with the hiring teams as well there is a

A Surprising Data Point: Why Vacancies Stay Open

a quarterly report issued by the Federal Statistical Office in Switzerland. They're sending out a survey to 80 ,000 companies on a quarterly basis, asking them about their activities, not only about recruitment, but it also includes recruitment -related questions.

And apparently 40 % of Swiss companies are reporting difficulties in filling open vacancies, which is kind of

of a paradox, since we have one of the best education system of this country, we have free access to a large, diverse, and well -educated talent pool. And yeah, we have more than sufficient salary premium

that we can offer to talents in Switzerland. So when they're talking about talent shortage, we basically felt a little bit uncomfortable with it.

and we decided to dig a little bit deeper and understand what exactly is the problem. 1We believe that there is no talent shortage in Switzerland. We believe the talent team failed to identify the talent.

Digging Deeper: What Recruiters and Candidates Actually Do

So we basically dig deeper, send out a survey. We talked to hundreds of recruiters and hundreds of job candidates and asked them what exactly do they do when it comes to hiring.

How do they prepare for the interviews? interviews, what questions they ask on the job interviews, etc. So obviously, there was a lot of feedback and a diverse amount of feedback, but we were

Three Root Causes Identified

able to narrow it down to three areas, which are basically here.

1) Noise: Too Many “Perfect” Applications

By far, I think by hiring team, the biggest problem that they reported is the noise.

So, like, there's a lot of buzz around hiring, and to be completely honest, AI and the recent investments didn't help at all. So hiring teams literally ending up with hundreds of CVs that are perfectly tailored for jobs,

2) Bias: When Teams Look for Shortcuts

which basically causes the second problem when the bias kicks in. So when things get complicated, yeah, usually the analytical part of people's brain checks out and bias checks in and everyone try to find the shortcut.

So if I need to compare it to a shortcut is basically when you walk down the aisle in the supermarket and choosing your toothpaste, there's no high stakes. They all look the same. You're going to choose one that's not too expensive, not too cheap. maybe the same brand as your toothbrush don't choose your talent like that right

so it's there's a much higher stake but what we saw basically when we were talking and interviewing these people they still looking for shortcuts they're looking for they make really biased decisions yeah and it causes a lot of

3) Candidate Anxiety: Talent That Doesn’t Show Up in Interviews

friction as well and on the candidate side I highlighted candidate anxiety that was by far the biggest problem reported by the other side of the hiring equation because there's a lot of talented people out there but they're not necessarily good in articulating their talent when it comes to a job

interview specifically because the stakes are much higher for those folks than compared to the hiring team so and when we were asking the hiring teams they basically confirmed that they rarely provide any type of help or or feedback for job candidates to showcase their best

or their true potential on a job interview. So there are a couple of hiring softwares and a couple of things that are addressing these challenges. We are working on this as well.

And I'd like to show you possible solutions for these problems.

Proposed Solutions: Clarity When There Is Chaos

Obviously, we like to create clarity when there is chaos.

Reducing Noise with Targeted Outreach

For hiring teams, basically, instead Instead of posting jobs on generic job boards like jobs .ch or LinkedIn or whatever, we like to offer targeted outreach.

So target those audience that you actually want to hire. If you want to hire an engineer, post your job or reach out to those engineers that can actually do the job. With this, you eliminate a little bit the noise.

Peer-Endorsed Skill Verification (Human in the Loop)

The other one that we are working on is peer endorsed skill verification, which is basically bringing back the human in the loop.

What does it mean?

So whenever you apply for a job, we basically ask the job candidates to select the skills and competencies that they have and ask also people that can endorse this with their social media link.

And we visualize in the hiring software for the companies in case there is a match, right? As you just see, it may maybe on the screen.

Reducing Bias Through Consistency and Competency-Based Assessment

The second is consistency to eliminate the bias. 1Competency -based assessments are extremely important, making sure that the hiring teams can carry out an easy decision but still have empirical data behind it.

So we like to offer them the possibility and automate certain tasks so they can do the toothpaste type of decision without data in the background.

Building Candidate Confidence with Automated Advocacy and Practice

And for confidence, we basically, what we're working on, what we didn't see with our competitors is that we automate the candidate advocacy.

So everything that's happening in the hiring software, in our tool, we basically take it and translate it for digestible learning modules for the candidates and they can practice for the upcoming interviews with AI.

So they pose with a typical interview question and then AI analyzes their answer and provide them feedback to improve and perform better on a job interview.

In this way, we basically help both sides create a fair, transparent, and more equitable hiring experience for job candidates and a smarter hiring decision for the hiring teams.

If you want, I would be more than happy to show you some demos about the tool, but if there's any question... Okay, which part are you interested in?

Q&A and Demo: Where AI CV Matching Breaks Down

So you might think that if everyone write a CV with AI and a motivation letter with AI, we also should use AI to analyze it, right? Which is okay, we actually implemented that, but there is a problem.

When Everyone Looks Like a 95% Match

if I go for example one of the open role is a project manager or product manager role that we have open you can end up with like people that 95 % match for the role hundreds of them right like it doesn't really help I think it's good if you have like an AI screening to screen CVs it's

great but if everyone sends the same CV and the same motivation letter most probably using the the same LLM, it doesn't really help. If you use another LLM to like analyze it and then everyone is a 95 % match.

So I can maybe show that one to you if I found the product manager.

So basically here are the applications that are coming in the tool on the date based on the applied, you click on AI match, it does the analyzes and create a short list for you on the left with the candidates.

These are the ones that usually should start with. it.

I think there's a big advancement because previous hiring softwares, they were analyzing only skill keyword matching. So they were comparing like how many keywords in the job description matches with keywords in a CV.

LLM goes a little bit behind it. It's great because it understands the nuances of how the CV is written.

So if I'm looking for a product manager, manager, product manager title may be not listed at all in the CV, but the LLM understand that there is a competency of someone that didn't list it in a CV, but it also there's a lot of friction and there's a lot of simulation between the lines as well.

So if I look at this like, yeah, good candidate, but I went through myself on these candidates and unfortunately none of them has the relevant skills although they match like 90 % and there are many of those yeah yeah so we have like we tested it

ourselves as well for this one I'm not ranked really well but for example I'm not a good product manager apparently but for a go -to market I never did in my life I'm apparently in a really good really good match I see yeah exactly so

that's why we implemented it we thought it's like oh yeah fight with fire against fire and then it doesn't really work so that's why we decided to

How Skill Endorsements Work in Practice

implement another thing bring back the human in the loop and have like a skill endorsement from former colleagues right so what we are working on now and the engineers promised is gonna be done in two weeks so hopefully it's gonna be done by May so that the skill endorsement so when the job application

the job applicants submit their applications and there is a match then And we're going to visualize it here as well on the candidates card, like how many skills actually match and how many were actually confirmed.

So what's happening is we offered the hiring teams as well to select skill tags. So AI analyzes the job attributes and offers you the interpersonal and technical skills that you can choose from. These are the ones that you like to measure on the job interview.

you and when this is not disclosed for the candidates and when the candidate submits the application we ask them as well like name the skills and competencies that you have and if there is a match in between this and the candidate submission which is confirmed by real people then we're going to visualize it in the card and say like it's a good indicator

that Peter built a hiring software who can confirm it I can put on a CV great everyone Everyone maybe does that, but like people that build it with me, they can confirm it and then if I'm going to come in the tool, they will see that actually I can really do that.

Yes. So no, no, it's good.

So when the job applicant sends the application, they select their skills, right? And they need to select the person.

and then what happens is that when you give your email address of your former boss it basically triggers an action we send an email to the email address that you have provided it's like mrs. Smith submitted the application for this role and it's she mentioned that she has this and this and this and these skills please confirm if it's right or not and then yeah it's basically yes yeah so

So it's like endorsing, really similarly what's been implemented in LinkedIn, but we're going to visualize it in a tool. But we send out an email, and then once it's confirmed and it's matched, we visualize it in a tool.

Closing and Thanks

Thank you. Thank you.