Startup Debate: Skills-Based Hiring w/ Maya Huber & Phuong Vu

Resumes are officially dead, and generative AI just blew the fragments to pieces. With a staggering 86% of hiring managers admitting that AI makes it too easy to fake skills on paper, corporate hiring has hit a massive signal crisis. The old way is broken—but what comes next?

The Face-Off: In this sharp, fast-paced episode of HR's most dangerous podcast, moderator Chad Sowash hosts a "Skills Duel", at HR Week's Fusion, between two visionary founders pushing the boundaries of talent acquisition:

• Phuong Vu (CEO & Founder of Telexa): The defender of the agile, skills-first organization. Phuong argues that true skills-based hiring isn't just about managing a massive, static library of keywords—it's about operationalizing human capability, looking for real-world evidence, and screening for highly adaptable, transferable skills.

  
  

• Dr. Maya Huber (CEO & Co-Founder of TaTiO): The champion of performance intelligence. Maya challenges the status quo, warning that "skills-based hiring" is on the verge of becoming an empty corporate buzzword if we keep relying on flat, retrospective data. Her radical solution? Ditch the applications entirely and force candidates to try out the job using live simulations before they ever reach an interviewer.

  
  

The Burning Questions:

• Are companies treating their employees like "disposable heroes" by reorging instead of upskilling?

• Has "culture fit" just become a lazy proxy for hiring people exactly like ourselves?

• How do you screen for a specific technical skill (like a CRM) when the technology evolves faster than the workforce can keep up?

  
  

The Verdict: Whether you are a VP of Talent Acquisition or a forward-thinking HR disruptor, this episode is a blueprint for the future of work. Stop asking candidates what they claim they can do, and learn how to evaluate how they actually think, adapt, and execute.

Listen now to find out who wins the duel.

PODCAST TRANSCRIPT (created by Gemini)

Chad: Well, thank you both for agreeing to have this debate. This is what we're calling the "Skills Duel." For all of you out there, my name is Chad Sowash. I'm the better-looking and smarter half of the Chad & Cheese, HR's most dangerous podcast, and I'm going to be your moderator today. That's right, this session needs a moderator.

Let me set this up. For decades, corporate hiring has relied on resumes: years of experience, job titles, and keyword-stuffed internal profiles. They tell us what someone claims they did, but not if they actually did the damn job. And guess what? Gen AI just blew that entire system to pieces. On the candidate side, resumes and cover letters can be polished to flawless perfection in three seconds. According to the data from Harris Poll, a staggering 86% of hiring managers say AI makes it too easy to exaggerate skills on a resume, and 80% say those resumes do not match real-world skills. Basically, everyone is lying, AI is helping them to do it, and traditional profiles are useless.

Okay, so now the landscape has been set. It's time to meet today's participants. In this corner, we have Maya Huber, CEO and co-founder of TaTiO. Dr. Maya Huber has a PhD and is an organizational scientist and performance intelligence pioneer. Maya is dedicated to replacing text-based hiring proxies with automated, real-world job simulations. She's an industry realist who believes a skill means nothing without specific job context and pressure to back it up.

And then in this corner, we have Phuong Vu, who is the CEO and founder of Telexa. She's a veteran workforce transformation strategist and has helped enterprise companies transition into agile, skills-first organizations. She is a forward-thinking disruptor who challenges corporate leadership to ditch obsolete job titles and scale practical, data-driven talent models.

Here's how the duel will run: each participant will have a two-minute opening statement. After the opening statement, we will start with the comment and rebuttal periods, both of which I will set as a baseline of about a minute for response and rebuttal, and deem more time necessary if I think it's interesting. So, let's get to the opening face-off. Phuong will be advocating for skills-based hiring, and her opening statement—are you ready, Phuong?

Phuong: Yeah, I'm ready.

Chad: All right. It starts now.

Phuong: So hi everyone, my name is Phuong, and I'm the founder of Telexa. Really happy to be here with Maya and Chad on this really interesting discussion. I honestly love the topic today, and I think skill has always been the controversial topics in HR and of course, in hiring specifically. And now with the rise of AI, the conversation has become even more interesting. AI is no longer just impacting like tasks or productivity; it starts to reshape the skills people need, how quickly those skills evolve, and even how organization think about like talent altogether. A lot of organization even struggle to find AI-relevant talent like right now.

But I think many organization today still treat skill as more of like a nice-to-have concept rather than like a true operating hiring model. And I think one of the biggest reasons for that is the complexity of applying that into real real-life, you know, hiring or like workforce planning. Because the traditional model is very straightforward—like head-count based, role-based, have you done this role before, how many years have you done that role before? But skills-based hiring model is significantly more complex because it require organization to understand evolving capabilities, the actual context of applying that skill, potential like let's say role and job opportunity for like one CV, and actually applying it to real business needs as well at a much deeper level. And I think that complexity is what make a lot of organization still hesitate to adopt the skills-based hiring model. It just way more complex and even require change of mindset and way of work in hiring and HR in general as well. From my point of view, like for me, as Chad said, I'm I'm very pro-skill. Skill obviously matters, they clearly do, and but I was I also do see real challenges with it, whether organization are ready or mature enough or maybe even...

Chad: All right, Phuong, that is your two minutes! We'll get into that though, we'll get into that. Maya, your two minutes start now.

Maya: Okay. So, first of all, great to be here, and I do want to cheer for a minute for the fact that we are two women founders here. Correct? Okay. So, I think, as you both know, the skills-based hiring is one of the important conversations right now, and my mission here is to make sure that this is not going to be another buzzword, like I feel for decades diversity inclusion was. So, as you just said, we have a signal problem. The market cannot rely anymore on CVs, on resumes, all those textual informations that were feeding our system. And this becoming a problem because the AI layer that organization are using to say or to claim that they are implementing skills-based hiring is is not full. It do not represent the full picture because those are old credentials or work history, or things that people have done in the past.

And what I'm intrigued with and where I think the future needs to go and real skills-based hiring is going to, is what is the next step of performance-based intelligence hire, where we analyze not old data, rather new data to understand where the market is shifting and what we are going to do with the skills of today tomorrow. Because the market is going to be completely different, positions are changing, and we need different infrastructure to assess our workforce. And by then to increase mobility, reskilling, upskilling, think this is real skills-based hiring, and as you said, Chad, for me skills without a context is useless. And how we can gather context in an era where everything we gathered so far is textual.

Chad: There you go, with time to spare. Very nice, very nice. Well, Phuong, right out of the gate, I'm going to go ahead and lend it over to you. And and Maya just talked about the buzzword of skills-based hiring, which I mean, DEI, it it seems to be coming into fashion, going out of fashion, and it has for over the last 20 years. How how do we actually operationalize all of this to make it where it's not a buzzword anymore? Because Microsoft has different skills than the next—every company feels like they're they're different. How do we operationalize? How do we provide structure? Help us out with that.

Phuong: So I think first of all, when we are looking at skill in general, I would say many company misunderstand what a true skills-based organization is supposed to be. [laughs] So, if skill is just treated as, let's say, giant skill library and everyone will be like, you know, "We have our own thing, we have this amazing skill library," or like map like these statics map mapping exercise, it's it's yeah, it's it's impressive, but it's actually not the real skills-based organization. Actually, I do agree with a part of a point from, you know, Maya is that, you know, skills-based organization is not just a skill library. If you are showing me like a skill library of thousands of skill, for me that is just skill inventory management.

So, I think a real skills-based organization require proper like in infrastructure. What I mean by that is it should like your skill framework should align with, you know, company culture, how you work, what are the product context, like your business goals, your actual business need, skill definition, actually clear proficiency and scoring framework, and also behavior indicator as well. Otherwise, like company end up building like these generic skill library that looks amazing and impressive but actually have no operational relevance. So, for me, it's it's really about that.

Chad: Okay. Maya?

Maya: So, my main point is we need to go deeper. Just tagging skills with, you know, on groups or identifying them is not enough anymore. Because the data we gather is flat. We know their work history. We and even if we are looking at employees' data, which I know you cover, Phuong, I think that the data that we have is their outcome. We haven't seen the path, we don't know their learning process, what failed in a way, what was not working. And looking at end results and bottom lines, both in textual information and employees' performance information, is becoming obsolete.

And I think the future should be there where we are leveraging technology right now to build a different infrastructure to assess performance, and by performance, I mean, putting employees and candidates within a position where we analyze their real actions, their behavior, how they solve solution within a context, specific context, you're right, in a specific job. But I feel and what that we are trying to do different things with the same data points, and this is why I'm afraid this is about to be a buzzword, because at the end, everybody will have the AI layer to assess their to tag their workforce with skills, right? That's easy. The the taxonomy is there. But how we can really know how we can take those employees to to different careers? And career is not linear anymore. We actually do not know how it's going to be, and I think this is a true calling for all HR people to do something else, not just adding AI-led layer on top of their performances, and have this feeling that they are doing performance, oh, skills-based hiring. It's not really.

Chad: So, real quick, Phuong, so we know that even when the market was moving slow—and it's moving incredibly fast right now—when the market was moving slow, the market had a problem keeping up with skills, right? And to be able to actually create structures and also training and those types of things. And right now, it kind of feels like before any of this can be deployed, it's going to be outdated. Can you talk about that? The vision of, okay, skills, we need to focus on skills for skills-based hiring. How do we ensure that we move just as fast as the market has is now, because we haven't in the past?

Phuong: So, I think that's actually aligned with a lot of organization now worrying about right now. So, a lot of them before like think about skills-based and they're like, "Oh, too complex to do." And then now they're like, "Oh, but AI also coming like, oh my god, do I have to do it now? Like too much work to do, for example." But from my point of view, like you don't have to like if you are trying to boil the ocean from day one, it's it's very difficult. So, from my point of view, like ideally, skill intelligence should eventually integrate into real business. And and that's why you can start small. What I mean by that is you can try by small experiment, like a much smarter approach will be maybe try a function, a department, align skills with that specific function goals and actually how the culture of that department works, identify a few capabilities that really related to the product that that that department or that function is trying to deliver, and then start building evidence like gradually.

Actually, a part of that I really agree with Maya is that, you know, for example, one skill, data analytics, for example. And you cannot ask like, "Have you used data analytics before?" like "Okay, I used that in the past ten years." But then actually, in reality, it's like, "Have you recently used data analytics to deliver a project in this organization?" So, that includes like delivery outcome, like how how peers and manager feedback on that skill, and business impact, and then collaboration effectiveness as well. So, I think the the small evidence could be gathering for that for let's say a small function. They can start start with small, so you don't need like a hundred skill map out for this department, but like focus on core skills. So, let's say if it's sales department, pin down like, "Okay, we need this year because we need to expand sales team. We need to focus on commercial acumen skill, customer engagement skill, consultative selling, and negotiation." And then for engineer, engineering, you start with like let's say four core skills. So, not every capability needs enterprise-wide like precision from day one. You can start with small and then stacking building evidence and focus on and focus on that. And honestly, I think that will be the shortest way to catch up with the world right now, rather than delaying. Because I think that if you delay any longer just because it's complex, it's it's probably just a bigger risk like long in the long term.

Chad: Well, Maya, I want you to answer the same question because creating simulations takes time too, and how will we not be outpaced by, you know, today's landscape, number one. And number two, please respond to, you know, what Phuong was just saying with regard to being able to create skills-based practices and why you think that methodology just won't work.

Maya: Yeah. So, first of all, simulations takes time, unless you're using the right tech. [laughs] Sorry, I can't ignore this comment, but but I will say, I I agree with Phuong on one thing: we should start small. But I think we cannot, the the way to build data is to build it completely differently. For me, the fact that we are asking people about something, they need to share their experience is wrong. Because first of all, we will not speak to every person many many hours, although I do see companies going back to this because they cannot rely on the resumes anymore, they need to talk to everyone, but that will take forever, and that will never work. And asking people about something is vague. They can say anything, resumes definitely can say anything.

I think we should put people inside situation, and if you are looking at organization right now, let's let's put hiring aside, recruiting aside. If I want to look at my organization right now and understand and unlock the the performance-based data that I have, I'm thinking how we can leverage technology right now, you know what, Chad, not even simulations, real gathering of data of their actual day-to-day at work, how do they process, what do they do, how do they communicate. Just tracking people, we don't need to go to the extreme, but even...

Chad: Well, this is after we've hired them though. What about before we hire them when we don't want to make the bad hire?

Maya: That's even, for me, it's basic. The way I think people should start companies should start right now is building a new data of of the candidates who come in and put them inside of scenarios, which is doable today, that are actionable. Again, not asking, not videos that ask people to share what they what they would do, not questions about what if, not interviews even. I'm I'm claiming that the fact that we are doing we move the recruiter to be an AI recruiter, it's great because speed matters. But that will predict the same signals that we currently have. So, going back to what you said, if an organization right now want to be ahead, not tomorrow, two years from now, with better data about their workforce, with better decision making, with the ability to to engage and make sure their candidates are equipped with everything they have, and their organization is sustainable, they need right now to to use the technology out there, simulation is one of them, job tryouts, any real-life scenarios out there that gather actions, performance, the way they're doing, the way they're thinking. Not just the end results, because everybody can share their projects. I'm thinking about how they were doing those projects.

Chad: So, Phuong, I'm going to I'm going to jump into something because I talk to VPs of talent acquisition all the time, and in many cases, without them even knowing, they they say, "Well, you know, from a training standpoint, we don't need to do that because the market has so many people that's out there, and we'll just go ahead and do a reorg and go find the talent that we need," right? So, it's like disposable heroes. You have all this great talent within the organization, and most companies are seeing that talent as disposable today. So, in both your models, Phuong you can go ahead and go first, in both your models, it doesn't seem like they care. They being the buyers, the talent acquisition, the people that are running the the talent organization, they actually care about the people and their skills, because they'll just go get more.

Phuong: Well, if I have to answer that short, it's it's going to be a hiring mistake [laughs] if you want to do that. I think, but also I think like for the skills-based hiring actually, I I would also link that with your previous question as well. Like at the start before the person going in, like from my my personal point of view, it's a com like when you're hiring when you hire someone, it's really like the skill that they have in what context of their experience, but also the culture fit. They could have the skill, they could have the experience, but do they actually work well with the current team? Like do they like will a person always use to the fast-pace of startup, or will do well in a large corporate like machine? For example.

Chad: Well, stop you there real quick because we've seen culture fit really just be a proxy for "people like me," right? And really just juice bias into the system. So, are we solving the wrong problem by even talking about culture fit when we should be thinking about hardline skills, in your case?

Phuong: I mean, culture like I think one more thing that people often see like culture is like, "Oh, someone like me." No, culture of an organization should evolve, actually. Yesterday is five, let's say, culture principle, today it should be six because the goals of the business actually already change. We need faster paces, therefore, we need better like actually higher bar of like autonomy in terms of people capability, for example. So, first of all, like culture should evolve. And the way that you interview should also evolve with that as well. So, within the hiring process, you should have and that's why even if you interview for a skill, like I would recommend to be like skill plus evidence. So, okay, data analytics, for example, like do you have that skill? Actually give me a specific example of how you utilize that skill in a specific context, like in a specific example. What the context of that project, what product you are trying to analyze, you know, in that in that example.

And also look at behavior signal, when I am like analyzing data, how do I communicate the data result like with other teams? Like what my manager said about my performance, you know, after I deliver that project using that skill. And also organizational relevance as well, what if I use data analytics skill and I come from a retail background, and then I'm applying for like a big data tech company, for example? So, all that is...

Chad: Okay, time. I got you on time. Maya, the disposable heroes question.

Maya: Yes. So, I'm I want to highlight something that I think is missed in the conversation about skills, and it is going back to what you said, Chad, about looking at the same people over all the time. We know right now that there are still although there are some industries who have enough candidates in their pipeline, there are still industries who are struggling to find qualified working hands, qualified ones. There are people out there with great skills who cannot find jobs, and this is part of the changing workforce we see right now, and we don't know where it's going.

What I want to say is that, when you are implementing performance-based hiring, and when you enable people to apply based on their actual current skills, and not what they've done in the past, again, skills, going back to what you said, Phuong, with context, with the culture of the organization, simulation to do that, but that's the way when you put only data on the front upfront, without asking people about what they've done in the before and give and asking people to give examples of things, just actual things tested right now, this is where you enable people from other experiences, other industries to to join the game. And this is crucial right now, and we see it on our day-to-day where when people suddenly say, "Well, I can do that. Actually, I know that, you know, I have skills that are able to do that." I'm always saying that if someone will look at my resumes, no one will say that this is a CEO.

And I and to your to your claim, Chad, I think this is a dramatical change we need to do. We need from the first moment to change the perspective we look at candidates, and to work with that. So, not start with a resume and then ask for those skills, start with the skills, start with performance, and then complete the overall data, and that everything related to to that so culture, organization fit, those are smart, but the infrastructure from the first signal should be different.

Chad: So, Phuong, when it comes down to skills, I can definitely put CRM on my on my my CV, right? The first CRM I used is Saleslogix, it doesn't exist anymore, and that was back in 1999, right? The things have changed, although that skill, skill, that tag, that's changed dramatically, right? How do you how do you actually how do you engineer for that? Because obviously, we have we've been in on the the years of the internet for for years now, for decades. These technologies have evolved, but they are so much different, and that's just one of the many different disciplines or skills that you could tag to. Talk a little bit about that because I think that's a huge problem today too.

Phuong: I think, so, [laughs] it's like when AI coming out and saying, "I'm looking for someone with like 10 years of experience." It's it's kind of like that. And for me, I mean, a skill on its self actually is not enough. I think what I would recommend organization to do is look at the broader capability. So, let's say if you have the skill like this skill using this CRM in the past, great. Tell me about a bit more about that of like when you're using that and at what scale you actually help your organization to use that CRM to deliver a value. That's one thing. But also, I care about your adaptability. That's another skill that is relevant to that. Okay, if I give you a new CRM today, can you adapt to it, and how fast you are adapting to it? For example.

Chad: Is this during the interview stage or is this during the screening stage? I mean because if you if you think about it all the way through, right? If this happens to the interview stage, I've just made it through the interview because it should have been screened out earlier, right? I think that's the hard part is how are we going to scale when we're seeing companies are getting tens of thousands of applications, most of them not qualified. How do we how do we get to the point where we're not interviewing individuals who shouldn't be in the room in the first place?

Phuong: I think, so I think this require change of mindset. So, if you are just looking for keyword of like HubSpot or like Salesforce, like right, you should not be hiring like that is not skills-based hiring. You like if you are look like you should be looking for transferable skill that are actually human-centric that will continue to evolve in the future. If I were the hiring manager, I would be looking for, "Okay, what like if they never use any system before," like that the red flag for me. But if they use some system before and over the time it looks like they use multiple different system, their adaptability, then I would question about adaptability skill rather than actually like, "Okay, how did you use that CRM?" I would not focus my interview based on that. So, it's a change of mindset and then is is the skill that they pick up to be transferable that think about human capability as unlimited, it can evolve, they can adapt, rather than like just specific on like a keyword. Like for me, that is just a keyword rather than a skills.

Chad: So, Maya, how would you deal with that? Again, this is a skill, it's been around for a long time, how how would you in your your system, your your ideology, how would you deal with that versus skills-based hiring?

Maya: My answer will be short because I know we're running out of time, but I will say that I think we should move forward, not for I agree with Phuong with different mindsets, but also different data gathering. Not what, rather than how processes, how people are doing things. The tech is not it doesn't matter, the CRM do not matter, the client they meet do not matter. What matters is how they're doing things, what's their set of mind, what is the processes, what is the decision making, how do they priori-prioritize their work, what how they think, how do they execute things, not the end outcome. And for that, we need to leverage the exciting tech out there to invite candidates to showcase their actionable skills, not subjective skills that they feel like that they have, other than that.

Chad: So, this is and again, trying to get to the where I was asking Phuong about where would this happen. Would this happen at the top of the funnel? Would they be testing, doing simulations? How does this actually work to make sure that I don't have unqualified people in the interview?

Maya: So, think of that: invite people to try out the job instead of apply.

Chad: Ah, okay.

Maya: So, the first thing they do is try out the job, they prove their actionable skills. If they like it, if they're good, they move into the next step, and they invite to the in to the interview. But we need to gather the performance data in advance to make sure, as you said, we are not investing our time in people that are not qualified to get the job done.

Chad: Excellent, ladies. Phuong, Maya, really appreciate you coming on having the debate, having the conversation. Again, this is for the the audience to to really understand what fits best in their culture, in their organization, and you know, for their workforce. So, really appreciate you guys coming on.