Transcript: ‘Codex for Nontechnical Builders’

The transcript of AI & I with Natalia Quintero, Every’s head of consulting, is below. Watch on X or YouTube, or listen on Spotify or Apple Podcasts.

Timestamps

Introduction: 00:01:05
How Natalia manages Claudie, the consulting team’s AI project manager: 00:02:35
Why the consulting team still pays for SaaS products: 00:04:55
Codex as a game changer: 00:11:47
Building personalized learning guides and illustrated explainers with AI: 00:14:55
Inside Natalia’s email triage app, and the ghostwriter that drafts replies in her voice: 00:21:40
The shift from knowledge work as sculpting to knowledge work as gardening: 00:26:44
Using Codex to one-shot a custom CRM: 00:28:57
Using Codex to build an app that coordinates her father’s medical care: 00:33:16

Transcript

(00:00:00)

Natalia Quintero

My favorite loop I’ve run so far on Codex was giving it a goal: set up my CRM to accurately reflect what had happened in my conversations and my inbox, across each of the hundreds of conversations with clients and prospective clients. I gave it a robust prompt and direction, and about six hours later, after I’d gone to sleep, it was complete. I woke up to a CRM that was fully set up—it had done what would have been weeks of work otherwise. It’s one of those moments of joy and delight with AI, where I wake up and my quality of life has improved.

Dan Shipper

Natalia, welcome to the show.

Natalia Quintero

Thanks, Dan. Good to be back.

Dan Shipper

So for people who missed your last episode, you are our head of consulting at Every.

Natalia Quintero

That’s right.

Dan Shipper

You are also the manager of Claudie, consulting’s AI agent employee, who was the star of our last episode together.

And I wanted to bring you on because I feel like every couple of months things shift so radically, and for me, you’re one of the bellwethers of how things are changing, because you’re an early adopter yourself, and you go and teach executives and other people at big companies how to use AI. So I think what you’re doing is a good window into how great operators and executives are starting to use this stuff.

The last time we chatted, Claudoe—the internal AI employee agent we built to help run the consulting business, to send out sales proposals and manage the CRM—was this nascent thing that Nityesh, our senior AI engineer, was sort of wizard-of-Oz-ing in the background, making it work minute by minute. But I feel like now Claudie is actually working. The model releases over the last couple of months have dramatically changed how much it’s able to do.

So give us an update on Claudie. How are things going there?

Natalia Quintero

You know, it’s funny—with the speed of AI, Claudie feels like it’s just not novel anymore. Claudie is an agent that does work for us every day. Claudie has its own LinkedIn and Twitter feed, manages our dashboards, and has a trust battery now—that’s new. It’s running on a loop to self-evaluate performance and improve itself given the feedback we give it. Claudie is thriving, I guess.

Dan Shipper

One of the things that’s interesting is you hired Claudeieto do operations work.

But you’re also now hiring an operations person. So what have you learned about the uses and limits of these kinds of internal agents for work you might otherwise hire a human for?

Natalia Quintero

Yeah, it’s really interesting. I think as we’ve all been using AI more, the thing we keep coming back to is that AI is really good at executing against a standard operating procedure, and Claudie is exceptional at that.

But Claudie still needs two things. One is constant oversight and management to make sure it’s actually doing those things well. The question of taste, of reaching for excellence, still requires direction and managerial support, which can be quite tedious and time-consuming, so there’s still quite a bit of time involved there.

And two: when you’re working with people—as much as I love working with Claudie, I want to interface with people.

Dan Shipper

I can’t really relate, but I see why someone might feel that way.

Natalia Quintero

And the reality is that while we have all of these rich dashboards and all of this data that Claudie is populating, we need someone to surface what’s interesting about that data, what the signals are, and to help lead those conversations. So I suspect we’ll continue to expand the team to build on the data and information Claudie surfaces, so we can actually do interesting things with it.

Dan Shipper

One of the big things you went through recently, which I think is super relevant to anyone inside a big organization or anyone running a software company, is you actually bought a CRM.

Previously, it was all Claudie glued together with Google Sheets. And there’s this whole narrative running around—I think honestly SaaS stocks are back, so maybe the narrative is a little less present than it used to be, but it’s still on people’s minds—of “are you just going to vibe code all your SaaS?” Fable is currently banned, but I’m sure it’ll be back, maybe even by the time this episode comes out. But if Fable can just one-shot a CMS, why would you use one? You have the ability to make your own CRM, and we have enough resources internally to vibe code one, but you decided not to—or you decided to move off the homemade one onto a professional one. Why would you do that?

Natalia Quintero

Yeah, so despite my hopes and aspirations that I could do all of these things and become an engineer and maintain all of these engineering products I’ve vibe coded—this one, I can relate to. It turns out there are private and public companies whose entire business is doing these things really well, sometimes very specific things really well. I vibe coded a CRM tool that let us manage our sales pipeline for a while—

Dan Shipper

—it was managing in Google Sheets, so Claudie was the glue between what was going on in Slack, in the meetings, and in Google Sheets.

Natalia Quintero

Yeah, exactly. Claudie had access to my email, could read our meeting note-taker’s notes, could digest any inbound leads that came in, and would track all of it in a Google Sheet. Eventually, that became a database we were managing.

And these things require maintenance. For the data quality to be good enough to do interesting things with it, you need to be on top of the quality of the data almost as much as Claudie does. It turns out this is Attio’s entire business. So I think one of the challenges with AI—I certainly have this—is that in this era, you can build anything. I think I even said this in the last podcast: the question is should you build and maintain whatever you actually build?

I think in this case, and probably in other use cases, we also rolled out Asana for our project management system. I think the scale of work we’re able to do because of Attio, because of Asana, and because of Claudie managing all of that information is much greater than if we didn’t have those tools. But now we have less burden on the team to maintain it.

Dan Shipper

Can you give me a concrete example? Because in my head I’m thinking, well, a CRM is just customer records, and that’s just a spreadsheet, so you should just be able to have Claudie do everything. Can you give me a deeper dive into what specific kinds of things came up that were harder than you expected?

Natalia Quintero

Yeah, so let’s talk about a traditional sales pipeline lead. They come in as an inbound. You have sales logic rules where certain things need to happen for them to move further down the pipeline until they’re a converted client, and sometimes those things happen very quickly, sometimes over a longer period. With my human brain, I can track what’s going on over a two- to three-month period. Any conversations happening outside of that, and after a certain amount of volume, I just can’t quite track.

With a tool like Attio, it has access to all the same things Claudie did, but it has really robust logic, so it can track the movement of a deal over the course of the pipeline and flag it to me in ways I would have had to supervise Claudie to do—and it wasn’t inherently set up for that. It could do it if I spent more time training it to do that, but it’s ultimately a reward-payoff thing.

Dan Shipper

I think one of the things that’s unintuitive about software is that real software is a compilation—a logical machine that compiles thousands and thousands of little logical rules that you wouldn’t expect you’d need beforehand. The whole job of the company and the engineers is to gather all the rules that are needed and put them into the system, and when something breaks, change the rules.

Natalia Quintero

Yeah.

(00:10:00)

Dan Shipper

And AI is very good at working around that kind of deterministic system and writing it, but it’s not going to one-shot all the rules you’re going to need.

Natalia Quintero

Needless to say, I’ve become a really big fan of PRDs.

Dan Shipper

Yeah.

Natalia Quintero

And scoping what I’m building—I think I’ve improved in both scoping and building higher-quality things, and also making the decision earlier about whether it’s going to be worth it to just invest in a tool versus build it out ourselves.

Dan Shipper

Yeah. I think a good metaphor—sorry, my brain is just cycling on the difference between software and language models—but a good metaphor is that software is a little bit like the bones in your body, and a language model is a little bit like your brain and your ligaments. If you didn’t have any bones, there’d be no structure, and you’d just be a flopping jellyfish on the floor. But if you didn’t have your brain and your nervous system and ligaments, you’d just be a pile of sticks. That’s a good way to think about how software and language models work together.

Of course, language models can grow bones, which is interesting—maybe a bit different from the way we’re set up—but growing bones well is complicated, and a whole body plan is very complicated. But you said something earlier that I think is really interesting and want to push on: I see you going from not technical to building stuff. Tell me if I’m wrong, but I feel like there’s been a step change in what you can build and what you can attempt over the last month or two. Do you feel like that’s right? If so, tell me more.

Natalia Quintero

Yeah, 100%. I’d say, riffing on Claude a little bit and the evolution of how I work with Claude and other tools—Codex has been maybe the single greatest improvement.

Dan Shipper

Never heard of it.

Natalia Quintero

I have to confess on the podcast that Dan did tell me to download Codex maybe every day he saw me for weeks.

Dan Shipper

I’m very annoying about things I think are good.

Natalia Quintero

And you have something I don’t have as much of, which is the fearlessness when it comes to trying out a new AI product. I still have a little bit of, “Okay, now I have to figure out this whole other thing.” I love Claude Code, and I’m very comfortable in the folder structures and file systems I’ve created. But Codex has been life-changing.

Dan Shipper

Totally life— so, thank you for your persistent follow-up.

Natalia Quintero

Thank you.

Dan Shipper

Anytime. Happy to be annoying anytime. Tell me why it’s been life-changing, especially coming from Claude Code or the Cowork universe. What were your expectations going in? What was it like? And how has it changed what you’re able to do?

Natalia Quintero

It really feels like Codex—I think you’ve said this before—looked at the things that weren’t quite working with Claude Code and then just fixed them when it launched the product. For me, having a non-technical background, having the terminal and the browser directly in the chat interface, and having such a powerful model as 5.5—you could just feel the compute. It wants to do hard work. It’s just so powerful.

I think over the past year I’ve gone through this transition of wanting to become more technical and trying to figure out what’s worth learning in order to do the things I want to build. I generally love learning—I’m an ambitious learner.

Dan Shipper

You really are. That’s something people should know: you’re the most curious learner I know. You spend your weekends having Claude or Codex build you these big learning guides that you read end to end about anything you’re thinking about. And I love it. I think it’s amazing, and it’s a superpower, because AI lets you do more of it and helps you use it better.

Natalia Quintero

I think it is a superpower, and sometimes it feels a little bit like a vice.

Dan Shipper

Yeah, yeah. Like, do I need to know the history of bookshelves from first principles?

Natalia Quintero

I would like to know that, yes.

Dan Shipper

—or something like that, because I feel like that’s something you would look up.

Natalia Quintero

I would like to know that. But with Codex, I feel like the truth is I don’t have to think so much about things like the file systems and the folder structures and how the scripts are set up—it just works. So I think I have to focus a little bit less on architecting things well, which is very much a skill, and something our engineers do extremely well, and just trust it to make good decisions and build solutions for me, which is really what I want.

Dan Shipper

So can you show us some of your Codex workflows?

Natalia Quintero

Yes. Okay, let’s see—let me share my screen here. We’ll start with… since we’re talking about learning, I have to show you my guiltiest pleasure, which is my favorite skill I’ve ever built. It was originally a prompt, maybe six months ago, and it codifies the way I like to learn things: what’s the history of a particular topic, what are the first principles that guide the physics of that topic, and then how did we get to where we are today, and what are the variables in the marketplace around it? That’s how I spend my weekends—reading these guides. And sometimes I don’t have 12 hours on a Saturday to read through them, so instead I make little cartoons that summarize what I’m seeing and learning.

This started out—if you can see my screen—as a prompt that Nityesh, one of our engineers, built, based on Claude, who works on the consulting team as the agent. Claude can basically teach principles and anything coming from this learning skill.

Dan Shipper

Go back up to the top. So tell me—what were you trying to learn, and how did this get made?

Natalia Quintero

One of the sad things that’s happened over the past six months is that as I’ve spent more and more 12-hour periods in front of my computer, I’ve prioritized my physical health less. So I’m trying to learn what I need to know to improve my physical education, and what I need to know to make more strategic workout decisions. I asked Claude to make a guide explaining the history of physical education—how did we find ourselves in a situation where we have to do specific types of mobility and workouts, and what do I need to know to make good decisions about how to spend my time on this topic? So Claude explains how we got to where we are—workouts as a topic, as an idea, emerged about 200 years ago.

Dan Shipper

Really?

Natalia Quintero

Yeah.

Dan Shipper

That’s earlier than I would have expected.

Natalia Quintero

Earlier?

Dan Shipper

Yeah, because I figured even 100 years ago, we were still doing a lot of physical labor.

Natalia Quintero

I think you’re right. It really became a thing during the Industrial Revolution, as people were spending more time in factories. With my learning skill, I could read all about that. But with the visual models Codex has, which are so powerful and so good, we could just make it a cartoon. This is something I could scroll through on the subway, or on a walk, or having coffee.

It takes these really complex concepts—one of the things I learned that was really interesting was anatomy, which I didn’t learn a ton of in school, and was really helpful to learn about. One of the most interesting things I really enjoyed from this set of cartoons was learning about the timescales with which different parts of your anatomy get strong—muscles get strong faster than ligaments, which get strong faster than bones, of course. Thinking about progression in physical strength as something that’s happening across your body, from your bones to your brain, as you said. This is one very fun example of something I’ll just do on the go.

(00:20:00)

Dan Shipper

How do you organize your Codex?

Natalia Quintero

Okay.

Dan Shipper

So you have a bunch of different projects. Do you use pinned?

Natalia Quintero

I only use pinned for my email triage app, which is the app you so generously gifted me this year. That’s the only thing I really pin. Everything else, I just work in.

Dan Shipper

Okay, got it.

Natalia Quintero

You’re a Codex pin person?

Dan Shipper

I’m a big pin guy, because I find that I lose stuff otherwise. I don’t have a project for everything, so all the work I’m doing is just pinned. But this is interesting—so you have a project for every sales strategy, your dad, and… “NZQ epistemology”? Incredible. Tell me more.

Natalia Quintero

These are my learning classes, I don’t know what to tell you. Suddenly really excited about how Aristotle came up with logistics systems, and how I use them today. We definitely don’t need to go into that, but—

Dan Shipper

It’s incredible. We actually might need to.

Natalia Quintero

I’ve been spending too much time around you. I basically just organize my Codex projects the way I would organize anything else. It does feel like—this is a thing I think Codex does really well—you had to do some mental organization in Claude Code, where you spent a bunch of time really understanding file systems and always had the finder open to understand where things were being saved and what was really being created. In Codex, that’s all happening in a really visual way, so there’s a little bit less mental load I have to take on. I just basically work in whatever project I’m prioritizing that day.

Dan Shipper

Okay, got it. And show us email triage—I’ve done a video on the way I use Inbox Sweep, or now we’re calling it Tend. This looks like you’re still using the original, but I think you’ve made some of your own custom modifications, which is another thing I love. I built an open-source app that lets you turn your emails into cards and blurs out anything you don’t want people to see. But this looks different from the app I made. So tell me how you use it, how it’s changed things for you, and what modifications you made.

Natalia Quintero

Yeah, so in V1 of the app you shared with me, it was obviously very custom to you, and it had these buttons to archive or send emails. There are a few things I need to do in my inbox—I’m either delegating something, tracking it in Asana, or… we work with clients that have hundreds of employees, and we need to track what’s going on across the different teams we’re supporting. So there’s a big mental load when I’m triaging my inbox, and I basically created a kind of second brain in my updated version of the inbox, because my email app can do a few things.

We can maybe blur out anything that shouldn’t be here. But as an example, my inbox was trained on this ghostwriter I built about a year ago—one of the first skills or prompts I built for myself. It’s trained on the most recent 150 emails I’ve sent, and it understands all the different contexts in which I need to communicate. Now it’s overlaid on my inbox—it has all the context of the work I’m doing across prospective and existing clients, and it drafts a note in my voice. There are a few things I can do: I can approve to send it, and it’ll get sent. We could ask it to rewrite—that was one of the great original buttons you had in your app. We could archive it if we don’t want to reply, or if it needs to go into its own markdown file.

Every client I work with has its own markdown file, and basically, at this point, my email knows what’s going on more than I do. So whatever it’s drafting is probably slightly more accurate than what I would have come up with. Sometimes I don’t need to reply—someone else might—but I do want that context to go into the markdown file. If I click the task button, it becomes an Asana task. I can send some things to spam, and there’s basically a save action here.

This is the kind of thing that’s just so insane, because you can build an app for yourself on the go. I was realizing I needed to triage my inbox and send stuff to different places, and I could just ask Codex to build a button that made that integration, and then keep using it on the go.

Dan Shipper

I remember we were sitting in the office on a Sunday, and you were making this extremely complex flowchart. Do you have that? Can you show the flowchart? That was a moment where I was like, “Holy shit, she gets it.” Bring up the flowchart, we want to see it.

Natalia Quintero

Let me see if I can pull it up. So, what you’re seeing here at a high level—can we zoom in a little more?—is a sales pipeline management flowchart. This is the kind of thing Attio just does really well: you import the logic, and then it can help you manage your pipeline at scale.

Dan Shipper

Is this for your email, or for Attio?

Natalia Quintero

This is for Attio. But it’s the same logic I need to use when I’m triaging my email—it goes to both places. When we get an inbound and it comes to my email, depending on whether it’s a fit for the work we do, there are different kinds of emails that need to be sent, and that advances as the conversation evolves. So this is the logic that enables me to do this, and it’s the same logic that enables Codex to do it.

Dan Shipper

You made this, and then how did you feed it into Codex?

Natalia Quintero

I PDFed it and shared it with Codex.

Dan Shipper

Sick. One thing that’s really interesting about this is what’s really hot right now is loops—everyone’s saying “loops,” but no one really knows what they are. This is an example of a loop. The way I’ve been thinking about it: previously, knowledge work—whether it was code, writing, email, whatever—was very similar to sculpting, where every single thing that happens on the sculpture is something you did with your hands.

I think knowledge work now is turning into something like gardening, where when you’re gardening, you’re creating the conditions for the growth to happen, but you’re not making the plant with your hands.

Natalia Quintero

Yeah.

Dan Shipper

And that’s what a loop is—instead of doing any individual email, you’re building the system that does your emails for you, and you’re intervening at different parts of the process. One of the things we talk about a lot is the human sandwich, at the beginning and the end, to say, “This is maybe worth my time,” and then refining the draft. And you’re trying to compound it—you create a flowchart, you do your email with that flowchart, which represents a loop, and every time you’re done, you compound the learnings back into the system so it gets better over time.

Natalia Quintero

Right.

Dan Shipper

Yeah.

Natalia Quintero

I think this really is an evolution of the model-manager analogy you shared, what, four years ago—where we’re going from using these systems effectively as individual contributors, asking them to do one thing or a small set of things really well, to creating a system, which is something a good manager does when they have a big team they need to help operate.

Dan Shipper

Couldn’t be me.

Natalia Quintero

Could not be me. But I’m glad that you’re able to do that.

Dan Shipper

It’s the same thing, right?

Natalia Quintero

Yeah, yeah, yeah. You need to create the conditions to help people succeed, and similarly, you need to create that shared context for AI.

Dan Shipper

Okay, so are there more things on your email app to show us?

Natalia Quintero

I think that might be it on the email app. There’s a bunch of other things I’m doing in Codex that I can share.

Dan Shipper

Yeah, show us some more, because just the email itself, I think, is life-changing—for you, and for me. I think we’re both getting through our emails way faster than we ever have before. Which is crazy. So what else?

Natalia Quintero

What else—so I can also share, maybe on the personal side… anecdotally, the best loop I’ve run is when we were setting up Attio. We were working with this really fantastic team helping us organize the logic of the CRM, and they asked me to enrich the information based on what had happened on the calls and in the emails. My favorite loop I’ve run so far on Codex was giving it a goal: set up my CRM to accurately reflect what had happened in my conversations and my inbox for each one of the hundreds of conversations with clients and prospective clients. I gave it a more robust prompt and direction to do that, and I think six hours later, after I went to sleep, it was complete. I woke up to effectively a CRM that was fully set up, and had done what would have been weeks of work otherwise.

That was only possible because of the shared context we’d created—it could make good decisions, good calls, with the context we’d built. It’s one of those moments of joy and delight with AI, where I wake up and my quality of life has improved as a result of this loop.

(00:30:00)

Dan Shipper

I guess before we move on from this—you do a lot of consulting, we do a lot of consulting, with executives at big companies, at tech companies, at hedge funds, and at PE firms. We do a lot of training of those people and their teams, trying to help organizations get more AI built like this and do work like this. So what’s the takeaway for someone like that who’s listening, about a workflow like this, and how they should think about whether and how to start incorporating some of this into their workday?

Natalia Quintero

I think my first tip would be to start with the systems you already have. If you’re already managing a big team, and you have KPIs and shared goals and OKRs that you’re tracking, give that same architecture or system you’re using to guide your team to AI, if that’s something your company allows. And then think about what tasks you want your people to focus on and do. At the end of the day, only I can get on calls and have productive conversations with our clients—for now.

Mike Taylor on my team did recently tell me he cloned me, so—

Dan Shipper

Exactly.

Natalia Quintero

—just remember me as the original version of Natalia.

Dan Shipper

How do we know you’re not already a clone? I don’t actually know.

Natalia Quintero

Well, we’ll never know. I might be a hallucination. So, yeah—start with that shared context, that shared infrastructure. Think about what you want only your people to do, and then start with small tasks.

I think the single biggest mistake I often still ambitiously make, and also see our clients make, is that you want to just remake the whole thing—you want to be AI-pilled, be AI-forward, be an AI-first organization. And so often, that just means you need to standardize and write down how you do a single thing really well. If you do that, and then the next task, and define what that looks like and when it’s done well, you can end up with more complex systems that can do sophisticated work for you. But the work, at its baseline, isn’t particularly sexy—it’s you having to read a markdown file, or a very simple set of instructions, and starting there.

Dan Shipper

So if you’re one of those people and you want to try a workflow like this—by the time this episode is out, we’ll have an open-source version of Tend, the email sweep app that Natalia just showed. We’ll put a link in the description. You can just throw it into Codex—or, honestly, you could throw this video into Codex, and it’ll just watch it and make something that works like it, but for you. But let’s keep going. I know you have some personal projects and other things you wanted to share.

Natalia Quintero

Sure. I’ll share—I’m personally fascinated by the role AI will have in how we run our lives. I don’t know if this is your experience, but certainly mine is that there’s just so much that needs to get done, and so many of those things are administrative tasks that I just can’t find time in the day to do.

One of the most recent things I asked Codex to do was to create an app that triages my dad’s care. My dad is 81, he’s the best, and he works with multiple nurses who support his care. There’s just a lot of health things that need to be triaged—medical appointments, follow-ups from recent procedures, and WhatsApp threads with me and with my family. So what Codex helped me do was basically create a kind of operating system for how, as a family, we could triage my dad’s care. This was a 13-hour project that Codex worked on, to go from a prototype to creating a full app that could help us with his care.

I’ll pull up the site here—it’s now a live app. What we’re seeing here is the portal my family shares for tracking what’s going on with my dad’s latest and greatest in his health. We get Google Form reports from the multiple nurses who support him, and we also have a WhatsApp thread with lots of casual updates about how an appointment went, or how his dosage on a certain medicine is going. What I have here is just a top-line summary—here’s the latest. I’m Colombian, so usually this is happening in Spanish, but if it’s the middle of the day and I need to know what’s going on quickly, I’ll toggle it and it’ll give it to me in English so I can digest it faster. But really, what we have is one central place—instead of having to dig through all of these different threads and sources of information, Codex has made it really easy to digest all of that in a single place, and to let us support my dad in what we can do best, which is to be present and loving as his family.

Dan Shipper

And your other family members are also accessing this? Are they also accessing it with Codex, or how does that work?

Natalia Quintero

No—this is just a password-protected website that we use and share. The nurses have a version of it too, so they can see what the other nurses have been working on, so there’s continuity in care. And you’ll love this, Dan—there’s a tracker for the different things each one of us is responsible for and should be following up on, which are things that, you know, we all have busy personal lives that we need to manage. Based on what’s going on in our conversations, these things get either highlighted as unresolved, or completed and grayed out.

Dan Shipper

I love it. So has this been amazing? What do the nurses think—are they like, “What is this?” Or, “This is the most organized family I’ve ever run into”? What are they thinking? Do they like it?

Natalia Quintero

It’s funny—I think a really good tool isn’t about the tool. I think the nurses just feel like we’re more proactive in showing up around the topics they need help with. So I think for them, it’s just that we’ve been better partners to them.

Dan Shipper

I love it. And it’s just one of those things where this is so obviously useful and good for you, your family, and for people, and I think that gets missed so often when we talk about AI being great at coding and stuff like that. It’s like—yeah, it is, and you can use it to do stuff like this, and people don’t realize how available and applicable it is to all the tasks we have to do, whether it’s caring for a family member or anything else in our lives. It takes a little bit off your plate.

Natalia Quintero

Yeah, definitely. I’m just so bullish on women using AI, and all of the administrative tasks that will suddenly be taken care of, because now we have this sort of super alien tool that can support on those things. I know Claire Vo has talked about that, who we love, and The Cut recently ran a big piece on how moms are using agents to do something similar. So I’m really excited about that space.

Dan Shipper

So one of the other things that’s happening for you is not only are you building these apps, but you’re building artifacts that help you learn things, for example, or just generally navigate the world. I think people think of AI as being able to generate text documents—slop text documents—but I think you’re using it in a way that helps with rich information transfer, which I think is really important. Can you show us some stuff?

Natalia Quintero

Yeah, sure. So maybe one example—I love Claude artifacts, they’re just so cool and powerful. One example recently is from a trip I took my mom on to New Orleans. Of course, the thing I was most excited to learn about was the pump system New Orleans uses, which is just incredible engineering—the kind of thing I don’t have time to do deep research into. So going into the trip, it was French Quarter Fest—Jazz Fest was the week after—and I created these artifacts on the go, as I came across things I was interested in seeing or learning about, and it would give us guides in Spanish so we could both share in what was interesting to us as we were walking around the city.

It would also—I asked it to read through my Spotify playlists to get a sense of what kind of music I liked, and then look at the lineup for French Quarter Fest—

Dan Shipper

That’s so cool.

Natalia Quintero

—and select which bands it thought we were most likely to want to see.

Dan Shipper

How’d it do?

Natalia Quintero

It was great—the timba and salsa bands were the ones that were highlighted, so we could use our time optimally to go explore New Orleans, and then when we showed up for French Quarter Fest, we could go see the bands that would most resonate with us.

Dan Shipper

I love it.

Natalia Quintero

Which just feels like a really fun use of AI.

(00:40:00)

Dan Shipper

Incredible. I love getting to talk to you—I always learn something when we chat. And if you want this kind of thinking inside your organization, Natalia runs our consulting. So if you want to get this out into your executive team, your product teams, and your engineering teams, reach out at every.to/consulting. And Natalia, we’ll have to do this again in a couple of months.

Natalia Quintero

Yeah, we will. Thanks for having me, Dan.

Dan Shipper

Thank you.

Dan Shipper is the cofounder and CEO of Every, where he writes the Chain of Thought column and hosts the podcast AI & I. You can follow him on X at @danshipper and on LinkedIn.