# Designing a Technical Workforce That Adapts at the Speed of AI - with Carrol Chang of Andela - **Show:** The AI in Business Podcast - **Published:** 19 Oct 2026 - **Source:** https://podcast.emerj.com/designing-a-technical-workforce-that-adapts-at-the-speed-of-ai-with-carrol-chang-of-andela - **Transcript:** https://podcasts.opennash.com/shows/ai-in-business/designing-a-technical-workforce-that-adapts-at-the-speed-of-ai-with-carrol-chang-845541/ - **Provider:** OpenNash Podcasts (machine-generated) ## In brief Andela CEO Carrol Chang argues that a job title no longer says what a technical job involves, because AI now does much of the hands-on work. Software engineering is the first profession to change this way. Companies should hire and assess for judgment, taste and problem solving, and plan their workforce by the tasks involved rather than by title. The main tension is that AI makes producing work cheap while making good judgment scarce. People without judgment will flood colleagues with AI-generated output, so hiring tests, including Andela's own, must separate a candidate's thinking from the AI's. Chang pairs this with a plan to map which jobs AI will hit first and to find early adopters. She expects work to become broken into tasks and spread across the globe. ## Key insights 1. **Same job title can mean very different work** Chang says some engineers now direct coding agents and review the output, while others still handwrite code. A title like 'full stack software engineer' therefore tells a hiring manager little about expected AI use. 2. **Judgment now matters more than coding syntax** Chang says AI reduces the value of knowing specific languages and raises the value of problem solving, strategic thinking and taste. Without those, people use AI to produce a lot of 'slop' that others must sort through. 3. **Hiring tests must expose the candidate's own thinking** Chang found take-home presentations so obviously AI-written that she could not see the candidate's reasoning. She is moving toward one-page memos and live discussion, and Andela scores copied agent output as zero. 4. **Map jobs by AI risk, then find champions** Chang recommends a top-down map of which job families AI will displace first, plus a bottom-up search for the roughly 5% of staff who are enthusiastic early adopters. Individual skill assessments should help workers, not monitor them, or trust and the rollout suffer. 5. **Bottlenecks will keep moving as AI spreads** Chang says engineering is no longer the bottleneck for some customers, so the constraint moves to product and then elsewhere. She also expects work to be split into tasks, some for agents and some for people anywhere in the world. ## Questions this episode answers ### Why don't job titles tell me what a software engineer does anymore? Carrol Chang says two people with the same title may work very differently: one directs AI coding agents and reviews their output, while another writes code by hand. Hiring managers need to spell out how much AI use they expect. ### What are 'squishy skills' and why do they matter now? Chang uses the term for skills between technical and people skills, such as problem solving, strategic thinking, judgment and taste. They matter more because AI handles much of the technical work, and these skills decide whether someone can direct it well and spot its weak points. ### How should companies test candidates when AI can do the take-home work? Chang suggests short written pieces, such as a one-page memo, followed by a live conversation about the candidate's reasoning. She also wants candidates to say which parts they used AI for, and Andela scores answers that simply copy an AI agent's output as zero. ### Where should a company start if it wants to retrain its workforce for AI? Chang recommends mapping job families to see which are most likely to be displaced by AI first, and starting reskilling there. She also says to find and promote the early adopters from every department and to offer hands-on learning. ### What will companies look like in five years if this goes well? Chang predicts AI workers handling repetitive tasks while people focus on relationships and strategy. She also expects teams that are more global, with work broken into pieces and sent to the best person anywhere. ## Facts - **Yolandi de Weerdt** [0:00:12]: "Andela connects global technologists with Fortune 500 companies to help them build high performing engineering teams for the AI era." - **Carol Chang** [0:02:37]: "software engineering is the profession and the job that Andela has specialized in for our entire decade plus history." - **Carol Chang** [0:03:46]: "rather than handwriting code, writing every character, what they are doing now is they are giving directions and guidance and guardrails and some context to AI, and then coding agents are executing the work." - **Carol Chang** [0:04:55]: "it's capable of doing the job at a breathtaking level of competency that you would have typically thought of an entry level worker needing to learn" - **Carol Chang** [0:06:29]: "There are hard skills, and these are the things that might sound like Python or Java or net" - **Carol Chang** [0:07:07]: "I've heard them called squishy skills before. These are things like problem solving, strategic thinking, judgment, and that bucket of skills is now at a tremendous premium." - **Carol Chang** [0:07:46]: "what AI is doing is it is making people's knowledge of all of the hard skills less important." - **Carol Chang** [0:11:38]: "We give them a prompt, we give them some data, and then we ask them to take a week and to come back and deliver a presentation." - **Carol Chang** [0:12:09]: "those five or ten pages are incredibly hard to read because so much of it is so obviously written by AI" - **Carol Chang** [0:13:10]: "we invested a lot over the last several years into assessment because we run this very large global talent network of engineers." - **Carol Chang** [0:13:46]: "people are gonna just go and put the prompt into Claude Code or Codex or GitHub Copilot and then take the outputs and and give them back to you" - **Carol Chang** [0:13:46]: "So we actually built a system in which we assume and encourage that people are gonna use AI, because that's actually how the job is changing anyway." - **Carol Chang** [0:14:21]: "we can spot that, and they'll effectively be scored a zero. They'll get no credit for that answer." - **Carol Chang** [0:16:06]: "usually then learning that hard skill is not the hard part." - **Carol Chang** [0:16:06]: "Andela is a company that was founded as a learning focused company." - **Carol Chang** [0:18:31]: "We believe in the seventy, twenty, ten learning model, which says that 70%, meaning the vast majority of one's learning comes from actually doing the work." - **Carol Chang** [0:19:23]: "once you unleash AI tools at a company, you will always have 5% or more of these folks who are enthusiastic early adopters." - **Carol Chang** [0:20:48]: "If you're assessing everyone individually in the workforce just because you're trying to do a monitoring exercise, right? That's not gonna win the trust of the workforce." - **Carol Chang** [0:21:59]: "It's gonna look like an org that has a lot of AI workers at the bottom, and those AI workers are gonna be in charge of a lot of rote, tedious, repeatable and automatable work" - **Carol Chang** [0:22:58]: "I had my best QBR ever, or I had my best QBR of my last year just recently. Because now engineering is no longer the bottleneck, and AI is what's enabling this." - **Carol Chang** [0:23:33]: "write a bunch of PRDs and ERDs, the product requirements documents and the engineering requirement documents." - **Carol Chang** [0:24:56]: "teams are gonna look a lot more global and a lot more distributed than they are today" ## Transcript **[0:00:12] Yolandi de Weerdt:** Welcome everyone to the Emerj AI in Business podcast. Today's guest is Carrol Chang, Chief Executive Officer at Andela. Andela connects global technologists with Fortune 500 companies to help them build high performing engineering teams for the AI era. In today's episode, Carrol discusses how AI is changing what technical talent needs to look like and why job titles alone no longer describe what a role actually requires. She walks through how leaders can redesign hiring and assessment to focus on judgment and problem solving, the skills that matter most once AI handles more of the hands on technical work. **[0:00:44] Yolandi de Weerdt:** And how mapping a workforce by task instead of title helps teams to stay flexible as rules keep changing. Today's episode is sponsored by Andela. If you offer AI products or services into the enterprise, you need to find enterprise leaders with relevance and readiness. Emerj attracts VP and higher ranking enterprise audiences who are already convinced that they need to move beyond traditional IT. To learn the exact strategies we use to help leading AI brands and startups connect with their ideal enterprise AI buyers, visit Emerj dot com. **[0:01:12] Yolandi de Weerdt:** Slash AD1. That's E M E R J dot com. Slash AD1. Now the conversation with Carrol. Carrol, welcome to the show. **[0:01:46] Carol Chang:** Thank you for having me, Yolandi. **[0:01:48] Yolandi de Weerdt:** I'm gonna start with something that is probably very simple, but just so that our audience members get where we're coming from with this conversation. Think job titles used to tell us a lot about a person's job and their skills and what they were expected to do. But currently, I'm not so sure that it still works like that. I think a role can look totally different than it did two years ago and same title, very different job description at the moment. **[0:02:14] Yolandi de Weerdt:** And I think the reality of that is that most companies haven't realized that they're still hiring and judging people the old way, thinking, you know, this was your title before at your previous company, so you would be a perfect fit for the title here too. I want to hear, get your opinion on that, but also figuring out how is AI reshaping technical talent needs, uh, role expectations, uh, hiring trends, those things. **[0:02:37] Carol Chang:** This is a great question to understand what is going on with AI and how big this shift is in knowledge work. I think that we should look at software engineering, and that's because software engineering is the first profession that AI has radically changed. AI will do this to probably every other field of knowledge work, but the first one that it has done it for is software engineering. And software engineering is the profession and the job that Andela has specialized in for our entire decade plus history. **[0:03:13] Carol Chang:** And so we see and we feel this, you know, more so than a company that might work in another, not field. And so what's going on is that you can have a job title, let's say a full stack software engineer, and that is that job title alone is not gonna tell you what that person is expected to do. Right now, there are certain engineers who have led the pack. They are super AI fluent, they are early adopters, and they are fully agentically coding. **[0:03:46] Carol Chang:** So what that means is that, rather than handwriting code, writing every character, what they are doing now is they are giving directions and guidance and guardrails and some context to AI, and then coding agents are executing the work. Agents are writing the code, the person is not anymore. And then the the engineer will go and review the output, give some feedback, ask the AI to give it another try. And that'll go back and forth a little while until there's finally a finished product that can be shipped into production. **[0:04:22] Carol Chang:** It could become live, and that is how some people are doing the job today, because AI has enabled that. The old version of doing software engineering is that you handwrite all the code. It's almost this artisan craft of creating your work product. And there is a lot of expertise that has to have built built in doing that. There's a lot of pride in that too. There's a lot of expertise and know-how. But AI came onto the scene, and most people in the world know AI for. **[0:04:55] Carol Chang:** You know, I could ask AI a question about my vacation planning, or, you know, some questions I have about, you know, what's going on in the world. But for actual knowledge work and for jobs like software engineering, AI is now capable of doing the job. And it's capable of doing the job, not at a senior level and not at a perfect level, but it's capable of doing the job at a breathtaking level of competency that you would have typically thought of an entry level worker needing to learn. **[0:05:27] Carol Chang:** And because of that, now, if you just see this job title, full stack software engineer, and you don't know whether the expectations are. We want you to do this job fully identically, and we want you to be totally fluent in knowing how to direct AI to do your job versus we want you to be not using AI at all. I actually don't know any engineering leader who is on that end of the spectrum anymore, but actually, where most of the world is right now is somewhere in the middle. **[0:05:57] Carol Chang:** Most engineering leaders are hiring for people who were trained traditionally, so they have some judgment and some know-how, but they also now are fluent in some of these cutting edge tools. And they don't want people to fully take their hands off of the wheel, but they want people who know how to use the tools. But people are at all different levels of fluency, and job titles alone are not going to tell you the way in which the job is expected to be done and the expectations for how much AI is supposed to be used in doing it. **[0:06:29] Carol Chang:** Very practical example, and I think for our audience members, this is gonna make a lot of sense from like leadership's perspective, where they would need to pinpoint the skills that this software engineer, modern day software engineer would have. How do we actually determine the skills? So obviously, if job titles don't tell us the full story story anymore, how do we actually know what someone can do? I think of skills as three broad buckets. There are hard skills, and these are the things that might sound like Python or Java or net, and these are, you know, generally coding languages or frameworks. **[0:07:07] Carol Chang:** And these are the kinds of things that you might think of as hard skills, meaning technical skills, and then there are soft skills, and we typically think of those as communication, teamwork, leadership, stakeholder management, cross functional communication. And then there is this bucket of skills in the middle that it sits somewhere in between them, and I've heard them called squishy skills before. These are things like problem solving, strategic thinking, judgment, and that bucket of skills is now at a tremendous premium. **[0:07:46] Carol Chang:** Because what AI is doing is it is making people's knowledge of all of the hard skills less important. It doesn't mean that you can totally give up on knowing coding languages, but what it means is that in order to fluently use AI, you actually need to possess those squishy skills. That is the bucket of skills that is most scarce and is most at a premium now. So judgment, taste, strategic thinking, these are the things that will separate somebody who is just using AI, frankly, too much and generating a lot of slop for their colleagues to read through, versus somebody who had really strong hard skills to begin with. **[0:08:35] Carol Chang:** And now they have superpowers, because now they know how to direct the AI and they know how to identify weak points in the AI's work product, how to shape it. And then in the end, they get to a work product that is the same or better quality that they would have produced in the past, but they get there much faster. And then they can take on a lot more work and still produce at that same or higher level of quality, but at light speed. **[0:09:01] Carol Chang:** And so the way that people are getting there is because they have the judgment, the strategic thinking and the taste. **[0:09:08] Host:** Yeah, and I I like the the bigger part of it as well. If we judge and hire based on skill, we would also know where the same person can fill different gaps within our organization. And I think that's the ideal idea or world that we we're thinking about. So let's say we have a leader listening to this and they agree with what we've been saying. Um, skills are changing, rules are changing, but they're stuck in the okay, so now what? **[0:09:35] Host:** We we can't fix an entire organizational team overnight. Um, and again, the reality is that most leaders don't really know what skills their team members have. How can leaders identify the first moves in building technical teams that stay flexible and productive as these new technologies Emerj? **[0:09:54] Carol Chang:** My advice is that people really should be placing a premium on assessing those squishy skills of technical taste, judgment, critical thinking, problem solving, strategic thinking. Because if you bring on team members who don't clear that bar, what's going to happen is you bring them on, you give them access to AI tools, they will produce a lot of work. And they're gonna create a lot of additional work for other people downstream, whether that's in code review or in reading a bunch of emails or updates that have been largely generated by AI but have not been filtered and winnowed down to the core message or problem. **[0:10:37] Carol Chang:** Really, when technical leaders or leaders at all of any knowledge work team are trying to think about how to build their team of the future, they've got to be thinking about how am I training for and assessing for those squishy skills of judgment, taste, critical thinking and problem solving. And I would suggest that when you have people going through an interview process, so I'm talking about when you are assessing people, new people to bring on to your team. **[0:11:08] Carol Chang:** There is a separate topic around what do you do with the workers who are already there on your team? But if the question is, how do I think about building a team? And how do I figure out how to find signal from noise and how to find the right people who are gonna get those superpowers with AI? You really have to focus on those squishy skills of critical thinking, problem solving, judgment and taste. And I would design parts of the assessment process that specifically hone in on that. **[0:11:38] Carol Chang:** I know, as an example, we always give people take home exercises and we call it a a final presentation. We give them a prompt, we give them some data, and then we ask them to take a week and to come back and deliver a presentation. And we treat it as a live conversation. So we wanna see how people are thinking on the fly. And I have found that probably in my last ten or twenty of these final presentations, it's overwhelming the amount of information that people are sending. **[0:12:09] Carol Chang:** We have said before in the past, you know, a five page or a ten page limit, depending on the role. And I'm now finding that those five or ten pages are incredibly hard to read because so much of it is so obviously written by AI that I cannot hone in on what that person's own thinking was relative to what AI produced. And so we're actually rethinking the way that we are asking people to produce work. I'm thinking, hey, just produce a one page memo where you are gonna talk about the most critical issues or assumptions. **[0:12:42] Carol Chang:** And then let's get on the line and let's talk through each of those things. I wanted to have a live discussion where I see your thinking process. Or if you're gonna use AI for the presentation, I wanna know what parts you used it for. Because that's gonna help me then understand what you got help with versus what you didn't. And then, of course, that relies on the candidate's honesty and forthrightness. But I think that assessment processes have to change and the way we think about assessment needs to change. **[0:13:10] Carol Chang:** At Andela, we invested a lot over the last several years into assessment because we run this very large global talent network of engineers. And our clients rely on us to find the right technologists at the right time for a certain job. And if you have a very large network, the only way that you can do that well is you have to have some technology based way of assessing and finding the right people. And what we have found is that in our technical assessments, just asking people to produce some code and not really giving them guidance or guardrails around how to use AI is not going to work. **[0:13:46] Carol Chang:** Because almost certainly, people are gonna just go and put the prompt into Claude Code or Codex or GitHub Copilot and then take the outputs and and give them back to you. So we actually built a system in which we assume and encourage that people are gonna use AI, because that's actually how the job is changing anyway. And so what we know, what the state of the art models can produce, and if people return to us in our assessments, basically exactly what the cloud code or Codex or GitHub Copilot or insert your coding agent of choices outputs were. **[0:14:21] Carol Chang:** And they don't do any additional thinking or problem solving on that, we can spot that, and they'll effectively be scored a zero. They'll get no credit for that answer. We are we design our assessments so that we are specifically looking for someone's critical thinking and their problem solving. And based on the path that that problem solving takes, then we know, okay, this is somebody who possesses excellent problem solving skills versus novice level skills. And it really is those skills that are at a premium now. **[0:14:49] Carol Chang:** And so I think that leaders have to design processes and at times rely on technology and tools too, that are really trying to hone in on those kinds of skills. **[0:15:01] Host:** It sounds like it's not really just the assessment process that is changing, but also the way that we will score the outcomes. So if we go, we do it this way and we have someone submit their one pager or their code or whatever, and our mission is to score the squishy skills. What do we do after that? Uh, okay, because we're no longer scoring their ability to code, we're scoring the squishy skills. So what happens? Do we then, if we identify someone with the right squishy skills, what happens after? **[0:15:29] Host:** Do we send them for training? Um, do we? How do we identify the gaps? What does that process look like? **[0:15:34] Carol Chang:** Certainly, if somebody possesses the right critical thinking and problem solving skills, but doesn't possess enough knowledge of the hard skills, meaning they don't know a certain coding language or they don't know how to use a certain AI tool, then usually filling that gap is much easier. And so at Andela, certainly, what we do is when we're, whether we're onboarding employees into our core team or whether we are onboarding contractors into our talent network, we will offer different ways to upskill to learn. **[0:16:06] Carol Chang:** You know, Andela is a company that was founded as a learning focused company. And so this idea of lifelong learning, continuous upskilling, you are never done with your learning journey has been core to us. And so we are always offering ways for people to learn that. If somebody has passed the right squishy skills bar and there's a certain hard skill they need to learn, usually then learning that hard skill is not the hard part. **[0:16:32] Host:** Ah, that's good news. I think that we have a lot of people listening to this and thinking, all right, it's my time to shine. I have loads of squishy skills just waiting to be harvested. So Carrol, I'm I'm gonna go a level deeper with our the final part of our conversation. And this is really where we get to give our audience members and the senior leaders some great guidance and advice. And the question here is, how can leaders leverage AI to understand work at a more granular level and build technical teams that flex as the needs change? **[0:17:02] Host:** That and they changing at rapid speed at the moment. **[0:17:04] Carol Chang:** I think that technical leaders, but actually, frankly, all company leaders are at this moment where they're wondering, why am I spending so much on AI? I don't know that I'm getting enough out of it. And what is my plan for my workforce from here? And I think that that is a very, very understandable, probably stressful place to be in. But for anyone listening who's wondering these things, I want them to know that everybody is asking the same questions at this moment. **[0:17:38] Carol Chang:** So you are not alone, you're in great company. And I think there's a really structured approach to handle this. I do think that leaders have to take a top down and a bottom up approach at the same time. One thing that we recommend is top down. You should do a map of your workforce. You should understand in different job families and roles that you have at your company, which kinds of jobs, just by nature of the job, probably are at risk of being displaced by AI first. **[0:18:07] Carol Chang:** And you should think about offering reskilling or some kind of pathway for those folks. There are other kinds of job families that are probably not gonna get touched by AI anytime soon. And so those ones you don't have to focus on reskilling first. And then for each of these job families that you do have to reskill, what is the learning outcome? What do you want people to be able to do in the new world? And how are you gonna give them the resources to do that? **[0:18:31] Carol Chang:** Whether that's providing traditional training programs or newer, untraditional training programs. At Andela, what we like to do is we don't like to do a lot of traditional classroom learning. We believe in the seventy, twenty, ten learning model, which says that 70%, meaning the vast majority of one's learning comes from actually doing the work. So we like to give people a lot of hands on practice, and that kind of thing can really help people develop those new skills. **[0:18:57] Carol Chang:** And I think that leaders have to do that. They have to map their workforce, they have to understand which of these job families they need to prioritize to get reskilled first. And after they have that heat map of their workforce, then they can create a plan. Because you, you get a list of which certain types of jobs and orgs probably need help the most, and you go in order. At the same time, I think that the company has to go bottom up. **[0:19:23] Carol Chang:** They have to understand from the individual level where people are, because once you unleash AI tools at a company, you will always have 5% or more of these folks who are enthusiastic early adopters. They come in and once they get their hands on the tool, they just shine. And these folks will come from all across the company. They're not just in your engineering org. They're gonna come from sales and marketing and finance and the people team. **[0:19:48] Carol Chang:** And those folks have to be identified and they should be celebrated. They should be showcased. You should make sure that those folks really get a platform to inspire other people to do the same thing. You know, those are the early adopters and the champions. And for everybody else, they probably do need some help, you know, getting along on this journey. And so what are the resources that the company is gonna offer them? There are gonna be some traditional training programs, some untraditional or newer programs, a lot of workshops and resources and show and tells, right? **[0:20:19] Carol Chang:** That you have to do. And and ultimately, at some point, I think it's a good idea for people to deploy individual level assessments for the workforce to understand their fluency in AI skills. We offer these kinds of assessments at Andela. I think they're super helpful in understanding where people's individual abilities are, but they have to be used carefully and respectfully. Because you want to use it to empower the worker, to have them as an individual understand where they stand and where their gaps are, and thus where they need to improve. **[0:20:48] Carol Chang:** If you're assessing everyone individually in the workforce just because you're trying to do a monitoring exercise, right? That's not gonna win the trust of the workforce. That is a recipe for making your AI transformation and your AI rollout backfire. And so there's all this advice I think that leaders need to have in terms of the do's and the don'ts, what's been effective and what hasn't. It's been enough time now that companies have been trying to roll out AI, that there are certain best practices that have emerged. **[0:21:14] Carol Chang:** And I think that companies need help and a little bit of handholding in that AI transformation. They need partners and they need tools and resources that are gonna help them through that journey. Yeah, uh, so much of what you said, just it sounds like. It sounds like a lot of work for now, but I'm sure that in the future it it will pay off. So I'm going to ask a final question. The picture that I want to leave our audience with today is okay. **[0:21:59] Carol Chang:** If we do this now, if we get it right now, if we change the way we think about hiring, assessment, training, the idea of a job title and a job description, what what is the pic? What does the picture look like five years from now? What is gonna be different if we get it right? Five years from now, the companies that get it right are gonna have a very different org structure than today. It's gonna look like an org that has a lot of AI workers at the bottom, and those AI workers are gonna be in charge of a lot of rote, tedious, repeatable and automatable work that a lot of the workforce actually doesn't even enjoy doing today. **[0:22:20] Carol Chang:** And then the humans are gonna be freed up to do a lot of the work that AI can't do, which is building relationships. This is stuff like client relationships, customer success, account management, strategic thinking and planning, and you're gonna see a lot more emphasis on those kinds of jobs. I think another thing that we'll see in five years is that the bottlenecks will have moved around. Engineering teams are so used to being called bottlenecks, right? There's like this endless backlog of projects that the business wants and the engineering team traditionally has struggled to keep up. **[0:22:58] Carol Chang:** That is changing now with AI. We have a customer who told us this really inspiring story recently, and she said that I had my best QBR ever, or I had my best QBR of my last year just recently. Because now engineering is no longer the bottleneck, and AI is what's enabling this. And so the bottleneck then moves, right? The bottleneck will move to product, right? So how do you ingest requirements from the business, create a plan, and that is then given to engineering that the engineering team can build? **[0:23:33] Carol Chang:** So the engineering team is building super fast, product becomes a bottleneck. Well, what if product no longer is the bottleneck? Because also AI is helping to do a lot of that business requirement gathering and to write a bunch of PRDs and ERDs, the product requirements documents and the engineering requirement documents. So if AI is helping to do all of this work, it's defining all of those specs, then where does the bottleneck go? Well, the bottleneck goes somewhere else in the business, and the bottleneck will keep on moving, depending on how fast AI can take on some of the rote work. **[0:24:05] Carol Chang:** And I think that in five years, you're gonna see these bottlenecks move way around, you'll see the bottlenecks move around like musical chairs, perhaps. But I think that the traditional notions of technology building being a bottleneck are no longer gonna be true. **[0:24:20] Host:** I think that is a very exciting prediction to look forward to. Uh, five years from now sound very exciting. I can see how these changes will help so many companies in different ways. When we're just thinking about workflows and work structure at the moment, but ultimately there will be cost reductions, there will be less staff turnover, so many other things will be positively influenced by this. So, yeah, definitely something that sounds very exciting and looking forward to the next five years. **[0:24:47] Host:** Carrol, it has been great having you on the show. Thank you so much for sharing your insights with us. I wanna have you back in five years and have this same discussion. **[0:24:56] Carol Chang:** We'll see if the predictions were true. There was one thing I wanted to add about the what'll happen in five years. I think that in five years, what'll also happen is that teams are gonna look a lot more global and a lot more distributed than they are today. And that is because AI is enabling access to knowledge work for everyone across the globe if they can get their hands on the right tools. Certainly, that access is an issue today and it has to be solved, and I think it is something that is on the radar of the big AI companies. **[0:25:31] Carol Chang:** But if we can solve that problem, and I am optimistic that we can, then this notion of I have a certain job I have to get done, you know, so I think about then. The traditional path is I'm gonna have to open up a job. Um, description and talk contact my recruiting team so that I can get a full time hire, right? But in C, I think that that is gonna radically change. I think that what AI is gonna enable us to do is to take that job, break it into pieces. **[0:25:59] Carol Chang:** Some pieces can be done by agents, some pieces of that work must be done by humans. And then how can I get that job out to the world and find the very best and most qualified person in the world to do that part of the job? I think that work is gonna get taskified, it's gonna get more fragmented, and it's gonna get more globally distributed. I do think that work is gonna be able to reach far corners of the globe in a way that it doesn't today in traditional hiring and recruiting cycles. **[0:26:28] Host:** Yeah, again, super exciting thinking about this. It it sounds like right now it's gonna take some effort to get to that point, but the result of the effort we put in today is going **[0:26:39] Yolandi de Weerdt:** Have an insane effect in so many different ways. But Carrol, thank you, it's been great having you on the show. I love the discussion and I think it's it's very optimistic and very positive. And I think it will also help a lot of listeners to understand that AI can be your friend, especially if you're in software engineering and feeling threatened by what AI is doing in your space. You still have value and you still have space, and you just need to make sure that you have squishy skills. **[0:27:04] Carol Chang:** Yes, thank you for having me. It was great to be here and see you in five years. **[0:27:23] Yolandi de Weerdt:** Wrapping up today's episode, I think there are three key takeaways from our conversation with Carrol. First, job titles alone no longer tell you what a role requires, since the same title can mean very different levels of AI fluency from one person to the next. Second, the skills that matter most now are judgment, critical thinking, and problem solving. These are the abilities that let someone direct AI well and catch its mistakes, not just the hard technical skills that AI increasingly handle on its own. **[0:27:50] Yolandi de Weerdt:** And finally, building a flexible technical team means mapping work by task rather than by title, so leaders know where reskilling is needed first and where people can take on new kinds of work as it changes. If you offer AI products or services into the enterprise, you need to find enterprise leaders with relevance and readiness. Emerj attracts VP and higher ranking enterprise audiences who are already convinced they need to move beyond traditional IT. To learn the exact strategies we use to help leading AI brands and startups connect with their ideal enterprise AI buyers, visit Emerj dot com. **[0:28:22] Yolandi de Weerdt:** Slash AD1. That's E M E R J dot com. Slash AD1. For further executive level analysis and to join our network of leaders delivering workflow impact with AI, visit Emerj dot com. On behalf of the team at Emerj, we'll see you on the next episode.