Ethical AI in Recruiting — Automating the ATS Without Losing Inclusivity

Introduction: What Encore HR Does

Thank you very much for being here. I'm Alessandro Razzi, I'm Sales Manager at Encore HR.

Defining ATS (Applicant Tracking System)

And before we start, does anyone know what is an ATS? No? Excellent. ATS stands for Applicant Tracking System.

So we develop what is the software that companies use to recruit people. So we manage everything from the creation of the job position, the multi posting of the job position because usually when you look for job you don't go directly to the company site you look through LinkedIn indeed or whatever whatever channel you want we had company to be there we had company to screen candidates set interviews and finally send offer letters if you ever

applied to companies like ADECO, Lamborghini, Aruba, Dallara or Saipem you don't know but you already was with Encore HR.

Today’s Focus: AI, Automation, and Inclusivity

Today we will see how AI can help company and especially HR department to empower and automate make repetitive tasks, but also improve the ethic way to use AI.

And especially I'm talking about inclusivity.

Demo Setup: Recruiter and Candidate Perspectives

So what we are going to see today is a little play. So I will be a full stack developer that is trying to apply to Encore at the same time, we will see what the

recruiter did in order to create this position.

Creating a Job Position with AI-Generated Descriptions

So right now, I I am already inside Encore HR, and I'm creating a job position. So everything I need is to write the name of the position, in this case, full stack developer.

If I click generate description, the AI will create this description of the job position.

This description is, of course, it knows what does a full stack developer, but the most important thing that Encore knows who you are. So know your values, know your company stack,

knows in what market you are in and it adapts the description based on all the information you already have in the backend.

AI-Optimized Inclusivity in Job Descriptions

I already talked about the importance for us for the ethics of AI and we can see here another magic button which is AI optimized inclusivity.

In this case our AI will analyze your job description because sometimes you want to use our regenerative AI but other times especially in big corporates you have to use what are the standard job description you have to use and this

tool allows you to see how inclusive are you because also the way we use words attracts or disincentivates people to apply to this position and this is critical for companies that really want to improve the way they position in terms of gender equality, of, generally speaking, equality of opportunities for people.

Candidate Application Experience

As a candidate, what I see is that on the left I have the job position and here I just have to start to fill basic information.

Collecting Candidate Data (Data Sets)

If I move on as a candidate, I will see the next step of the hiring process, that is what we call a data set. Right here, they just ask me to add my resume, my number, my address, to move forward.

From the recruiter side, of course, I created this data set because this is the data I want to track of candidates. Because probably in the hypothesis that you don't fit this position, I can use your data for future position to ask you to join the company.

So it's important that I track as many information as I want in the in this case I just want to know curriculum mobile number and residence address, but I can add as many data set as I want

Self-Assessment Questions Generated by AI

moving forward we let's say Encore wants to have a very detailed hydrogen process

So what we have here we want our candidates to self assess their knowledge about various aspects about the job position So this question was actually generated using AI, so AI knows what are the skills you need in order to succeed in this position, because of course it knows the position and it knows the company.

So here it asked me to self -evaluate me in various aspects, like how proficient are you using React, because the software knows that my stack is React. So I have to be honest on answering this question. Do you have experience in integrating AI and machine learning technologies?

We are in core. We are seeing how we use AI so Let's say I'm proficient how familiar with git for version control and so on and so on

So it helps both the candidate and the recruiter to assess the the level of the candidate Okay, perfect.

Optional Step: Video Interviews for Soft Skills

Moving on, we can add a video interview. We are not going to see it today because unfortunately we don't have time, but we also have a video interview where the candidate should answer questions that will appear on the screen.

This is important for many reasons, but the most important one, the one for which our customer uses, is to see the soft skills of candidates even before setting an interview

with the HR department okay perfect so this was the recruiter experience in

Recruiter Screening and Evaluation

encore but of course I applied to a position and now I want I want to see how it looks like to screen candidate which is usually is the second step in HR department so I create this opposition I multi posting in the various cat channel I wait for the candidates to apply for the position and now I want to

screen I can see here on this position that I have many candidates that applied for so here you can see the basic anagraphic details of the candidates here I can see the scored question so the self evaluation of the candidate which is something I can start using for screen candidates to understand who

could be a good fit for the company and here on the right we have the two AI tools of Encore.

Keyword Matching (First-Generation AI)

The first one is the keyword matching. It's something that has been around for I would say at least five years so it's basically what we are looking forward is to search on the candidate resume certain keywords

for us could be important for example react if I see react it will appear in this in this table or other words that are important or I can name competitors to understand probably they work in the same area where I work on and could be

interesting to see where so this is the first generation of AI I'm not spending

AI Ranking: Scoring Candidate Fit

much time here because what is more interesting today is to see the ranking AI so this tools allows me to see how much this resume of the candidates match

with the job position it's a score between 0 and 100 no one will have have 100 out of 100 because the perfect candidate for the job position, it doesn't exist. You can ask any HR in the world, they will say the same thing.

So there are candidates that are a good fit and candidates that could be a good fit if trained properly. So from here, we can see which resumes are much closer to the job description.

Explaining the Score: AI Resume Summaries

And if I move with the mouse here, it doesn't give me only the score. It also tells me why it's a good fit or not. But if we want to have a deeper analysis of that, I enter in the candidate page, I click here, and here I have what the AI elaborated about the resume. Probably you are not familiar with it, but it's very much similar of what recruiting agencies send to final customers to companies when they present resumes.

Reducing Bias: Removing Gender, Age, and Geography Signals

1So the first thing that AI does when it has to evaluate a resume is to erase completely the data about gender, about age, about also geography. Because the AI don't want to know your ethnicity, your gender, and your age. It just wants to know how could you be a good fit for the company.

And that's very important because sometimes humans and humans, if I'm searching for a junior, sometimes an HR would think that if I'm 40, I could not be a good fit. But maybe I was a waiter for many years, I started with a course or I just graduated because, you know, life never knows what it has in your path and I'm ready to commit for a junior position.

So what it does, it helps also the recruiter not to have biases but to have a powerful sum up of your resume. me. For every position, what he analyzes is different.

For example, for here, we are looking for competencies on the technical side, integration with AI. But of course, if I'm looking, I don't know, for a construction worker, I won't have this field.

The only thing I have here is the conclusion that is always present on all candidates, where the AI tells me why could be a good fit and if it's not a good fit what should be what should I dive into during

the interview to understand if it could be a good fit and let's imagine that I want to hire this person even though the AI says it's not a good fit what are the area where I have to invest best in training in order to put him at the same level of the other colleagues.

Workflow Automation: Candidate Communication and Anti-Ghosting

From here of course we can see also the resume, we can send emails directly from the platform to candidates in order to keep them aligned on the journey and of course where we see

here the orange dots it means that we have automation because one thing I don't know if you are into the HR world but one thing that HR and also candidate hates and I agree with them is ghosting so this automation prevents ghosting

because every time you move a candidate from a state of the pipeline to the next that could be also the let's say not interesting candidates we you can send a communication to candidates so they you can keep in touch with candidates that you are interested in and also be polite to the ones that don't reflect what you

are searching for so I am a little bit early on the on the on the agenda so if If you have questions, the Q &A, okay. Okay, we can see so more.

Collaborating with Hiring Managers and Protecting Data

Another very important thing when it comes to recruiting process is when I have to involve the line manager. Because in companies, you have the HR department that search for profiles, but of course, then they have to work with the software development team or the production team.

Sharing Candidate Profiles with Controlled Visibility

and so sharing the candidates in the right way also to line manager is crucial we have this feature the sharing profile where i can share the profile directly from the platform but if i want I can select what to share with the line manager.

to hide to the line manager details about i don't know um the the resume of the candidates because I don't want him to be biased by facts like age or gender or the mobile number of the candidates because I don't want him to have access to personal data etc.

Maybe I can want only to share the AI analysis of the CV because I know that is completely safe or just some notes I am putting here to help the line manager to better understand the candidates and he will receive a link where he can consult what I want him to to see in in a streaming way so you

GDPR and Keeping Candidate Data Secure

don't have CVS that goes around on emails and documents that could be a problem when it comes then to GDP our policies because of course if a candidates want to get back all the data they gave you you must be sure that the data of the candidate is not in your system so this helps you also to keep

safe data of candidates all right that's it

Finished reading?