Episode Transcript
[00:00:06] Okay, everyone, today I'm going to give you a little bit of a vision into the future, and I'm going to paint you a picture of how we can use AI tools like Galileo to completely automate almost the entire process of talent management. So let's start at the beginning. You're a hiring manager, maybe running a sales team, and you want an open requisition for a new staff member. So you go to the AI agent and you say, I want to open a requisition for a new hire. And the agent says, okay, tell me about the job, the level, the title, what are the responsibilities? Explain the goals, what are these skills you're looking for, some of the experiences you're looking for, and give me a synopsis of everything that you're hoping to find in this person. The AI records that conversation, prompts the hiring manager to answer all the questions and stores it. Then the AI says, okay, would you like me to interview anybody else on the team? And you say, yeah, why don't you interview Brody over there, who's one of our top salespeople, and interview Josh, who's also one of our top salespeople, and then interview sue, who's our head of client service, and ask them what they believe the criteria are for success and what they're looking for in the next sales rep. And the AI goes out and has a conversation with those three people because it knows who they are and collects that information into its corpus. So now it has a corpus of three or four interviews to define this job. Then it looks at the practices for hiring in the company, which we've given it, the hiring levels. And it looks out on the Internet and scans for jobs and roles and job titles that do this kind of work in different companies in our industry, in the city or cities in which we're hiring. And it comes back and says to the hiring manager, here's a prototype of the job description, here's the level, here's the recommended pay, and here's the description as you and your team described it and the information I got from the outside market. And of course, it now knows the various competitive positions out there and can make your job description better. The hiring manager looks at it and says, okay, that's more or less correct, but I want to make sure you have the company values and the company corporate culture in there. And so the AI looks at the company mission statement, the leadership model, maybe a few CE speeches, and says, okay, I've tweaked the job description. I've added the employment brand and we've got it where we like it. Next thing the AI says is, would you like to source internal candidates? And the manager says, yes, I'd like to consider that, but let's look at the external market first. So the AI goes out into the market and starts looking for pools of candidates. And by the way, it also looks at internal candidates and it evaluates the candidates and scores them based on this complex corpus of information it has. It has the corporate values, it has the business objectives of this job, it has the job levels and job titles inside the company, it has the input from the two peers, it has the input from the manager and has the input from the outside market. So we're not creating a job description for which there are no people. And it starts to source candidates. And when it brings the internal candidates in, it can do other things to learn about those candidates, which I'll talk about in a minute, but let's just talk about the external candidates now. As it sources external candidates, it goes through a series of criteria, such as, where are they located? Are they willing to work full time, Are they willing to come into the office? How long is the commute, what is their current pay level? Do we have people coming from that company into our company regularly? Is this a strong source of hire or a weak source of hire? What is this person's experience? Where have they worked before and so forth. And so we can train the AI to be quite sophisticated in its scoring. And so it brings back, say, three to five candidates to the hiring manager or the business partner, by the way, and says, here's the top five people I've found. I have a variety of criteria that you can sort them. You can sort them by experience, you can sort them by culture, you can sort them by leadership potential, you can sort them by relative pay and sort them by culture, et cetera. And so the hiring manager working with HR goes to the list and says, maybe there's 10 people on the list, I want to interview the top three based on these criteria. And the AI says, great, let me set up the interviews. The AI sends an email to those three people and says, congratulations, you've been shortlisted. We would like to set up an interview. When are you available? And it finds the interview available times from the candidate comes back to the hiring manager and the hiring team. And by the way, you can tell the AI who's on the hiring team. And it says, I'm going to set an interview rubric for you. Would you like a. How long of an interview? Would you like to have and how many questions would you like me to create? And the manager says, how about 10 questions? And we'll divide them up amongst the following four interviewers. So each one will have two to three questions to dig into and then provide me five additional questions. The AI goes and builds a set of behavioral interview questions. And you say to it, I would like a rubric for answers. So the AI creates behavioral interview questions as well as a rubric that describes what a good answer would be like and what a poor answer would be like. Produces it, sends it to either the business partner or the manager. The manager looks at it and says, that's great. Question number seven, I don't really like, let's come up with a new one. Goes back and forth. We come up with the hiring guidelines, the interview guidelines. Then the manager says to the AI, I want you to give these questions to Sally, I want to give these questions to Brody, I want you to give these questions to Josh, and I will handle these others and I want the HR team to handle these. We set up the interview process. The candidate is available on these hours. And the AI says okay, let me go out and schedule these interviews. The AI schedules the interviews, sends emails back and forth to the various candidates and interviewers. The interviews take place, all the interviews are recorded. The AI is monitoring all the interviews. Now the AI has not only a corpus of requirements, but now it has a corpus of candidate information and a corpus of interviews. And in each interview it can evaluate the candidates based on the criteria set in the rubric to give you scoring, as well as direct feedback on the candidate's experience, level, maturity, intelligence, perhaps level of culture, fit and other criteria based on your needs. And you can see that what this machine has now done is almost the entire process of what the recruiting function has to do with multiple steps of people. But it goes on. So let's suppose the manager says, I like Joe. We all agree on him. We're going to send him an offer. Here's the offer, here's how much we'd like to pay him. Send me the offer letter to review. The AI shows you the offer letter. The offer letter is reproved. It goes out to the candidate. The candidate takes it, has a few questions. The AI goes out and does a background check. The AI says let me do some references. Sends an emails to a couple of their references to get check on them to make sure that their background is valid, then comes back and says, when would you like to start? Goes back to the hiring manager and says he's Available to start in two weeks. We pick a start date. The AI builds an onboarding guide because the AI has the standard onboarding guide available that you've done for other candidates. But they know that this person has a particularly different background. So we're going to set up a slightly different onboarding process. In addition to going through the standard onboarding process, we're going to have them spend some time with a couple of unique people because they have interests in a particular area. So we set up interviews for that. The person joins the company and gets started and they go into their role in sales and all of that is still done by AI and all of this data is captured. Now three months go by and the AI looks at the performance of this person, looks at their sales performance, looks at their revenue performance, looks at their feedback if their there is any, looks at their behavioral performance. And by the way, if you use products like the ones that are coming out now from various vendors, it can actually look at the sales calls that it's made. And it goes back to the manager and says Joe was a good hire. But based on the three month performance of Joe, based on other people who started in the last year, he is slightly behind on the curve. I did an analysis and I found that if you look at the correlation as to why he's behind, it seems to have to do with the fact that he did not work for a company that has ever funneled into our company before, or he has an academic background or a slight difference in his sales experience. So that could be the reason he's behind. I wanted to give you some coaching on some things that might help. Would you like me to factor this into the future hires for new criteria? The manager goes, yes. So now the AI has learned from Joe's first three months on the job. In the future, as we select and source candidates, we should be a little more discriminating in these particular areas. This process goes on. Joe gets more effective. He starts to sell very well. He's highly successful. Six months, nine months a year, two years. Turns out he was a great candidate. He's being considered for promotion. The manager goes back to the AI and says, we're considering Joe for a sales leadership position. Can you give me a sense of what you think would be key success factors or blind spots for Joe? The AI looks at Joe's sales calls and various recorded meetings and comes back to the manager and says based on other sales leaders in the company, based on data that we have about our leadership model and data that we've been provided by you. Here's some of his blind spots. So if you're going to prepare him for leadership, here's some things to think about. By the way, I'm happy to coach him if you'd like. So the AI says we know a lot about Joe, we know a lot about sales. If Joe would like a coach, I will be happy to coach him. Now, I left out pay, I left out questions about benefits, I left out questions about salary, questions about quota. All these other things could all be done by AI as well. And I'm really just starting to think about this now. The reason that I wanted to record this is that we just finished a very in depth exercise in the company here to build a document with 100 use cases for Galileo and HR. And when you look at them and you read them and they'll be available soon, what you're going to see is they're far, far, far more sophisticated than crafting a job description from a bunch of words. They're very sophisticated, data driven, analytic, end to end agentic processes. And I know that this is where this is going. Now the scenario that I just walked you through is pretty sophisticated. But I'm telling you, there's vendors right now that are moving in this direction. Seekout, Eightfold, LinkedIn and others. I don't know all of them that are out there. And you can see that if you consider the agent or the AI actor as an end to end system, the data that is collected across the talent life cycle is extremely valuable from process to process to process. Now this gets into systemic HR and the whole concepts of systemic hr. We didn't talk about diversity, we didn't talk about pay equity, but all those other things could be done in the same fashion. And so my firm belief, as we move into the era of super workers in our companies, which you're going to read more about soon, is that this type of end to end approach is going to be the way we're going to run hr. And there is an analogous scenario for learning and development that's very, very similar to what I just talked to you about in talent management and recruiting. I didn't talk about the internal hiring process, but of course in the internal hiring scenario you would want to probably check with the current manager and get information from the current manager about somebody's strengths and weaknesses to build a development plan for this person when they start a new job. And there's many, many, many more examples of this. I don't think this is going to be difficult to do in 2025. And so I want you to unleash your imagination about what you can do with tools like Galileo. Now, let me get back to Galileo for a minute. Galileo is a very, very intelligent agent. Not only is it trained on the 25 years of research and case studies and best practices and vendors from us, but it is built on a platform that uses a router to route your questions to multiple large language models in a highly intelligent way. And there's features in Galileo to store content like meetings and recordings and documents and interview guides and your own internal information to build an agent exactly like the one that I just explained to you. It also has features to parallelize processing. So if you're for example, looking at interview analysis of 10 candidates at the same time, you can say to Galileo, compare these 10 candidates based on their fit with our culture, and here's a document that describes our culture. And you can see Galileo visually analyze the 10 candidates at the same time and show you the relative strengths and weaknesses of each. And you could do the same thing for technical skills, you could do the same thing for pay, and on and on and on. So these systems are extremely powerful and they're going to have a massive positive impact on our jobs in hr. The reason I keep pitching Galileo is not because we're trying to become an HR tech vendor, but I do believe that a lot of what's missing in HR right now is simply awareness and imagination of what these things are capable of doing. Now, many, many vendors are working on many, many creative solutions and we're going to have lots and lots of options to pick from in the coming year. But I suggest you start with Galileo as fast as you can because it's priced very modestly and get your hands dirty with these tools and build these kind of processes yourself because the vendors can't possibly consider all of the scenarios you're going to want to do in your company. I hope this is a good introduction to the entire end to end agentic view of HR going forward. And I'll continue to put these kinds of scenarios together for you to help you as an HR professional, as a leader, as a vendor, or as a consultant. Thank you.
