How fast can you upskill in AI? We did a sprint to find out.

JAI PUNJABI: Can we just take a minute to acknowledge what everyone’s done? You can pat yourself on the back or do whatever it is you need to do to reflect on where we started a couple of days ago and how much progress we’ve made, both in terms of our skill sets and what we’re capable of doing.

RANA EL KALIOUBY: That’s Jai Punjabi, co-founder of WaitWhat, the company that produces this podcast and other shows like Masters of Scale and Rapid Response. Jai is speaking to a Zoom room of every member of his team. And what these tiny Zoom squares are now capable of is something that does deserve a pat on the back, because a couple of days ago his team were AI novices, some even AI skeptics.

MG: I’ve been a pretty staunch AI hater, I must confess. And I feel like I have gone through a major transformation. I think these three days have been amazing, and seeing the way other people have fully revolutionized their workflow, I’m truly so blown away by what people have created. So I’m mostly just impressed.

RACHEL ISHIKAWA: MG is a video editor at WaitWhat. Going from AI hater to appreciator in three days’ time, that was by design.

EL KALIOUBY: For years now, we’ve heard about the productivity gains we’ll achieve with AI, from people like Microsoft AI CEO Mustafa Suleyman.

MUSTAFA SULEYMAN: This AI moment is going to deliver the greatest boost to productivity in the history of our species in the next couple of decades.

And if there’s one thing certain about AI, it’s that it changes everything about how we work. It can feel like a tsunami with no timeline. And every company is asking a variation of the same question: How should we be using AI?

ISHIKAWA: Our company set out on a mission to find out. What happens when a company closes normal operations for three days and has every single member of its team test the limits of AI? Well, on this special episode, we’re going to tell you. I’m Rachel Ishikawa, by the way, senior producer over here at WaitWhat.

EL KALIOUBY: And I’m Rana el Kaliouby, and this is Pioneers of AI.

[THEME MUSIC]

Copy LinkWhy companies need an AI wake up call now

EL KALIOUBY: So let’s start with this. Tell us a bit about what you do and your background.

PARTH PATIL: In short, I’d say I spend 14 hours a day talking to language models.

EL KALIOUBY: Okay.

Parth Patil is an AI engineer and one of investor and serial entrepreneur Reid Hoffman’s AI advisers. And if you want to know how to integrate AI into your work, he’s your guy. Which is exactly why WaitWhat brought him on as an AI consultant for this whole experiment.

ISHIKAWA: Three years ago, after he saw what ChatGPT was capable of, Parth quit his data analyst job, cashed out his 401(k), and decided to spend every waking moment dedicated to AI.

PATIL: I was just like, oh my God, even if this doesn’t get better, we’re going to change the world with this. If it does get better, then all of engineering changes and all of a sudden becomes accessible through natural language.

EL KALIOUBY: What is special about this AI moment compared to previous moments, say, the industrial revolution or the internet?

PATIL: This is the first time we have a computer that can use language and that can speak. The idea that you can wield a computer through natural language means that you kind of have a steam engine for knowledge work. It reminds me of Harry Potter and spell-casting. If you know the right combination of words, things start happening. That’s kind of what I feel like we’re at right now.

EL KALIOUBY: And poof, just like that, Parth turned nearly every aspect of his life over to AI: his work, his finances, his email. It’s not something he recommends for everyone.

PATIL: If you give it access to your email, you’ve already given it the keys to your kingdom. I only recommend it to people who are willing to lose sleep over the security situation of these things.

EL KALIOUBY: Look, you don’t have to turn over the keys to your kingdom. Personally, I’m concerned about the security risks, so I’m keeping email to myself for now. But if you’re running a company, you do need to be using AI strategically.

PATIL: What’s at stake is that you’re going to be competing with other new entities that are AI-native and move in ways you didn’t think were possible. If you don’t get ahead of it, you won’t even realize when it’s already happening. I’ve seen this again and again, where the AI-native player is secretly eating the market share of the legacy players in a space. You shouldn’t let that happen to yourself. You should create an environment where you evolve out of that position.

ISHIKAWA: And this mindset, to get ahead of AI before getting lost, was a huge motivating factor for WaitWhat to upskill our team.

PATIL: It was on a Saturday. Taryn reached out, Taryn Fixel.

TARYN FIXEL: My name is Taryn Fixel. I’m the president and COO of WaitWhat.

PATIL: She reached out, and she met me at the studio that I work out of because she wanted to learn from me how I used some of my favorite tools. She basically looked at me after two hours and was just like, do you think you could teach a whole team how to think like this in maybe two days?

FIXEL: And I suggested that we pause company operations for three days in order to all align around how we’re going to use AI and what we can use it for.

PATIL: Honestly, I was very skeptical. I know that if I sit next to a friend for four hours, I can definitely AI-pill them, and they’ll be changed moving forward. But I was like, how do you do this for a whole team of people who I don’t know that well, but who have a strong intrinsic motivation to learn? So that’s, I guess, the most important thing.

EL KALIOUBY: And there was a strong intrinsic motivation.

FIXEL: If this were Blockbuster 10 years ago, 15 years ago, wouldn’t you have wished you had done this? Imagine Bob Iger is in this role. Would he do this? So we made the decision to make a bold move, pause company operations fully for three days, and all explore together.

Copy LinkHow a companywide AI sprint was designed to solve real bottlenecks

ISHIKAWA: So the sprint was off. Parth, along with WaitWhat’s leadership, devised a three-day experiment. The WaitWhat team was divided into 12 smaller groups to tackle bottleneck problems. Some of these problems were related to the content we produce. But there are other important parts of our company. For example, we need sponsors to make this content. We also hold an annual conference called the Masters of Scale Summit.

EL KALIOUBY: And a plug here for you to attend, this October 20 to 22.

ISHIKAWA: Yes, definitely come out. But as amazing as Summit is, and as amazing as all of our content is, it takes a lot of work to pull off. And AI can help us solve some of our pain points. So these 12 groups set out to solve problems like this.

KELSIE SAISON: How to automate our episodic assets that we put out weekly.

ALEX MORRIS: Trying to think about clip generation.

CHLOE GOSHAY: The creation of partner deck materials.

EL KALIOUBY: On Parth’s recommendation, the teams focused on two main tools, Claude Code and Replit. Teams had an additional budget of $100 to experiment with other AI-powered software.

ISHIKAWA: And just to put it in perspective, pausing operations for three days for a production company isn’t a small thing. No interviews, no research, no episode production. It meant that we needed to drop reruns for a week and also play a lot of catch-up after the sprint.

Copy LinkWhat hands on experimentation with AI actually feels like

EL KALIOUBY: And there’s the financial element, right? You have to invest in these tools for your team to use. OK, so the ground was all set. But what was the sprint actually like? Because I was just an observer. You actually experienced it.

ISHIKAWA: It was good and hard and frustrating. Each group was on Zoom pretty much all day, screen-sharing as someone took the reins with Claude. Because right now these tools are single-player. There isn’t a whole lot of collaboration that can happen within the app. And like we’ve talked about a lot on Pioneers, latency is a real part of using AI. You ask it a question and then you have to wait, sometimes a good bit of time. And so we’d fill in the gaps.

LEITAL MOLAD: I hate Gatorade.

STEPHANIE STERN: If you water it down, it’s better.

MOLAD: I find Vitamin Water a lot more palatable than Gatorade. We were trying to figure out how to use AI to help with guest curation for podcast interviews and live events. This is literally a full-time job.

And the group was doing all the right things, having these Socratic conversations with Claude. We were all using a speech-to-text tool called Wispr Flow so we could just talk to Claude directly, which is what Leital was doing before we started building.

MOLAD: Can you ask us some questions so you can get a better idea of the mission of our company and how we typically select guests and speakers for our podcasts and events?

ISHIKAWA: But like a lot of the groups, they also encountered some technical hiccups. At one point Claude Desktop stopped working.

MOLAD: Claude Desktop failing to open for some users.

STERN: Oh, interesting.

ISHIKAWA: Please update to Claude Desktop v1.0.1, blah, blah, blah. But where they landed on day three was pretty cool.

DORCÉ: So what did we build? We built a standalone webpage. Ooh, this is sexy, right? It has some really good information on here. And Stephanie, please chime in. Eve, Leital, chime in when you see fit. What it does is it took speakers that we’ve had previously and other speakers, and it created new suggestions for us. It tagged whether the speaker was on Masters of Scale, Rapid Response, or Pioneers of AI, or if it’s suggesting those speakers for those particular shows.

ISHIKAWA: So picture a dashboard with different tags and guest suggestions. There’s a little context about each guest, too.

Copy LinkWhere AI works best with structured data and operational workflows

EL KALIOUBY: But how good were these suggestions?

ISHIKAWA: Some of the guest suggestions were almost too obvious. I mean, who doesn’t love Oprah? But there were some lesser-known names that the tool also identified.

EL KALIOUBY: This is really cool. We’re always trying to find the people we don’t often hear from to feature on Pioneers of AI.

ISHIKAWA: There’s still a lot of tweaking that needs to happen, but I think overall the team is happy with where this is headed. I think it’s worth noting that this was one of a handful of projects for the sprint that relied on AI generating new material. But I’ve got to say, some of the more successful projects were driven by objective data, projects like the one DeAngela worked on.

DEANGELA NAPIER: Hi, I am DeAngela Napier, and I am the special events project manager for the Masters of Scale Summit. I’m an independent contractor that has been working with WaitWhat for the past three summits, and I’m here again to help them with Masters of Scale Summit 2026.

ISHIKAWA: Again, Summit is our big annual conference here at WaitWhat. DeAngela doesn’t need convincing about the perks of AI. She’s been using it for her own creative practice. But for her job, before this whole experiment, she mostly used AI as a thought partner. She saw the sprint as a chance to work more efficiently.

NAPIER: Before we started this sprint, I really just thought about all the things that I do. What’s something that could be better? Because after four years of doing something, you sort of have systems, but then there are a couple of little pain points where you’re like, I wish this was better. So I was thinking about hotel management, because that’s a big part of what I do as it gets closer to Summit.

ISHIKAWA: Hotel management may not seem like it, but this is a huge task. It involved keeping track of every speaker’s travel information, what hotel they’re staying at, their personal preferences, like what floor they want to stay on. And this information isn’t static. Travel plans change a lot.

EL KALIOUBY: I am totally guilty of that!

ISHIKAWA: A lot of people are. And DeAngela is the one who has to keep track of all of this information coming from all these different avenues, from Slack, email, sometimes a phone call. For a while she was using Google Drive to keep track. But after three days with Claude and then Replit, she created a new Hotel Operations Dashboard. Basically, using a system of standardized forms, the dashboard updates live. She demoed this for the whole company, and it was definitely a fan favorite.

NAPIER: I have to say that I was actually surprised that people thought it was so amazing, because I just thought, well, I’m just helping in my role. But then as people were talking to me about it, I just thought, yeah, I could really find other ways to apply this to other people’s roles that could help them. Because once you have one system, you can repeat it.

ISHIKAWA: I think people were also impressed just by the sheer amount of work that goes into hotel management. I mean, I was. And after DeAngela demoed her project, Taryn, COO and president, made that clear.

FIXEL: There is so much invisible work in everyday roles. It is very easy from the outside to look at somebody’s role and go, what does this person actually do all day? And the job description does not fully capture the reality. I think this is where AI introduces a really meaningful shift in creating workflows that are observable, explainable, and shareable.

DEANGELA: It was nice to hear because the people in our department know the things that I do, but I always get the feeling that most people don’t. I come from a military family, and my dad was like, it doesn’t matter what accolades you get, just do the job well. And so I’m going to do it well anyway. But it’s nice to have that recognition.

Copy LinkHow to face the fear that AI may reshape jobs

ISHIKAWA: Look, DeAngela deserves all the accolades. But I feel like we have to address the elephant in the room. There was this tension swirling beneath the entire sprint.

EL KALIOUBY: Say more.

ISHIKAWA: The whole time we’re working on this sprint, there’s this fear that maybe we’re not just learning how to use these AI tools, maybe we’re training these tools to do our jobs. We all remember those headlines from Dario Amodei about how AI will have major impacts on jobs.

EL KALIOUBY: Sure, but my fundamental belief is that AI in the long run will add more jobs into the economy.

ISHIKAWA: Right, and you’re not alone. Jensen Huang, president and CEO of NVIDIA, among others, has said something similar.

HUANG: A lot of people say AI’s coming, we’re going to run out of work, our jobs. It’s exactly the opposite. The fact of the matter is PCs made us more busy. The internet made us more busy. Mobile devices made us super busy.

ISHIKAWA: But speaking personally, as someone who was on this sprint journey, Claude can do a lot of the things I do as a producer, for example, research, which is a big part of my job. And to put a point on this, there used to be an associate producer for this show, as you know. And while AI isn’t necessarily the catalyst for removing that position, I am now using AI to do some of the work an AP typically would do. And I will say, no one at WaitWhat is trying to hide this tension, especially leadership.

PUNJABI: I can imagine there isn’t a single person on our team, or really on any other media company team, who’s not looking at the headlines of jobs being cut, things being replaced, being asked to do more with less.

ISHIKAWA: Again, Jai Punjabi, co-founder of WaitWhat.

PUNJABI: And so you pause everything for three days and you send everyone into a questioning mindset of, what does this mean for my job? What does this mean for our role? What does it mean for the future of the company? And you’re kind of figuring those answers out as you go. You’re not sure.

ISHIKAWA: Jai said that he knows this isn’t the most satisfying answer, but leadership’s response to these fears is to be as transparent as possible and to involve the whole company so it doesn’t just feel like AI is happening to us, but like it’s something we’re actively shaping together.

FIXEL: Every business will be disrupted in the next week, year, five years. There is no industry that this will not impact.

ISHIKAWA: Again, Taryn, COO and president.

FIXEL: And I want to, number one, make sure that the organization is nimble and has the ability to thrive in the future media landscape that we live in. And number two, I actually really do feel that we have a responsibility to every member of our team to give them access to these tools and help them feel empowered to use them.

EL KALIOUBY: Right, so it’s not just upskilling the team to be better for the company, it’s also about developing a team you care about on an individual level.

ISHIKAWA: That’s how leadership sees it.

EL KALIOUBY: Interesting, because that’s how Parth sees it, too. But he takes it a step further.

What’s at stake if organizations don’t upscale and reskill their employees and their teams?

PATIL: I like to start with people and think about, OK, agnostic of whether you work at this company, what’s at stake for you is that you could learn faster, you could build your brand, you could be more resilient in your career. You could learn anything. If you can unlock that at an individual level, then it’s OK if the current job doesn’t make it. I joke to people, a lot of companies aren’t going to make it. But that doesn’t mean your learning isn’t going to make it.

EL KALIOUBY: Not going to make it. Right, right.

For organizations as a whole, the stakes are just as high. AI won’t just replace your job, it’ll replace your entire company.

PATIL: And what’s at stake is if you don’t become AI-native, anyone on your team with high potential is not likely to stick around long term. If they do stick around, it might be because they’re like, oh, I automated my job and no one knows, and then I can just coast. For a lot of people, they may go build your competitor and beat you because they’re able to go deeper and further and faster because of these tools. And what’s at stake is that you’re going to be competing with other new entities that are AI-native and move in ways that you didn’t think were possible.

ISHIKAWA: Can you imagine if I just started my own AI podcast? Kidding!

Copy LinkWhy the biggest outcome may be a mindset shift

EL KALIOUBY: OK, so maybe you’re not going to start your own podcast, and it doesn’t seem like team members are trying to spin off with a competing media company. So what are the results of the sprint?

ISHIKAWA: There are concrete AI tools coming out of this sprint, and we can talk more about that. But I do think it’s important to point out the less tangible outcome: our mindset shift. And I know that can sound kind of contrived, but a lot of us left the AI sprint in a different headspace than where we began.

MG: I did not use AI before the sprint. There are a few different reasons.

ISHIKAWA: Remember our AI hater? That’s MG, video editor extraordinaire.

MG: Number one, as an artist, I’m also a writer and a performer. These are industries in which the artist oftentimes doesn’t have as much agency and is taken advantage of. I feel bad for artists whose work has been scraped from the internet, copied, pushed through this sausage-maker AI thing.

And I think environmental concerns are a big one for me, just worrying about all of the processing power and how that is affecting our planet, and specifically places in the Global South and places that are not necessarily in the U.S., not where I live, but disproportionately affecting other communities.

EL KALIOUBY: Those are some huge concerns about AI. These are fundamental issues that we’ve talked about at length on this podcast. Was it that simple? It just took this sprint and those concerns disappeared?

ISHIKAWA: Definitely not. MG still doesn’t use AI in their personal life, and they still don’t think AI should be used for everything. But they did have a clear mindset shift.

MG: I have to say, the last day of the AI sprint, every team got to present their work, and it kind of blew me away. The tools that people built, I was genuinely like, OK, so I work with just a ton of geniuses. Crazy.

ISHIKAWA: For the sprint, MG was working on how to implement AI to get to a rough cut of video episodes faster. And honestly, they did not find a solution for that particular workflow. Even so, they saw the impact AI could have in other areas of work.

MG: I’m very impressed, and I’m very pro people using AI in ways that take away the hateful tasks. My mom and I will say, if we have to do something like taxes, then we have to go to the Ministry of Hateful Tasks. So using AI to minimize your time spent at the Ministry of Hateful Tasks, great. I love it.

Copy LinkWhat it takes to turn promising AI experiments into daily practice

EL KALIOUBY: Wow, I love that. OK, so we saw changes to mindset, but what happens next? What about the more tangible outcomes of the sprint, like DeAngela’s hotel management system, Leital’s guest suggestion engine?

ISHIKAWA: Truthfully, a lot of that is still being worked out. DeAngela’s app still needs more tinkering before it can fully be integrated into her workflow. And it’s the same story for the guest engine.

EL KALIOUBY: And there were 12 projects, right? It would actually take a lot of time and resources to get all of those projects off the ground.

ISHIKAWA: Exactly. Right now there’s a super AI tiger team trying to figure out which of the projects from the sprint will have the biggest impact.

FIXEL: It is not enough to simply do a three-day pause if we’re not then taking the learnings and applying them to our day-to-day workflow. The point of doing a three-day pause is to apply it to your day-to-day workflow. We’re still figuring that out. What are all of the challenges that we believe we can only solve with AI that help substantially push forward our KPIs?

ISHIKAWA: And that process is going to take a lot longer than three days.

EL KALIOUBY: It’s like the sprint can get you 70% up the mountain, but then that last 30% to the summit is steep.

ISHIKAWA: Yeah, Jai describes it like this.

PUNJABI: Everything feels very imperfect right now. With each step forward, it’s kind of dark. You can see feet in front of you, but then you keep stepping and come to an intersection, and you’re like, maybe I’m going left, maybe I’m going right, maybe I’m not going down that path altogether. But I think you still kind of have to walk down the path. It feels like trying to figure it out together and see where you can go.

ISHIKAWA: So Rana, I think that’s where we leave it. There’s no tidy bow.

EL KALIOUBY: But honestly, that seems like where we’re at with AI. There is so much possible now, but we still have to figure out what actually sticks.