Claude's new code interpreter tested live against ChatGPT's agent mode across Excel, PowerPoint, and pivot tables
Nate B Jones and his guest Rod test Claude's newly launched code interpreter feature in real workflows, comparing it directly to ChatGPT's agent mode.
Summary
Nate B Jones and Rod, an entrepreneur with an extensive AI background, conduct a live comparison of Claude's newly launched code interpreter against ChatGPT's agent mode. The central finding is that Claude consistently delivers polished, usable outputs — multi-tab Excel spreadsheets with working formulas, well-designed PowerPoint slides, and heat-mapped pivot tables — while ChatGPT's agent mode produces more ambitious plans but fails on execution, delivering broken spreadsheets and poorly laid out presentations. Nate also demonstrates a workflow using Perplexity to generate self-contained, data-rich prompts that can be fed directly to Claude, bypassing the need for Claude to do its own research. The episode closes with a live exercise building a movie night pivot table, which Claude returns with heat mapping, spark lines, and friend-level viewing statistics that were never requested.
Key Takeaways
FULL TRANSCRIPT
Introduction and Context
Nate B Jones: All right, I'm on here with Rod. Rod, do you want to just introduce yourself and tell the audience a little bit about you and what makes you interested in AI?
Rod: Yeah, sure. Thanks for having me on, first of all. Great to be here. Basically, right now I'm an entrepreneur. I started my own company about two and a half years ago and I've used AI extensively through that process. It's not my first experience with AI — when I did use it before, we used it around things like third-party risk intelligence, where you're trying to understand these very nuanced scenarios that ladder up to these more complex situations that you need to de-risk. And then two and a half years ago when ChatGPT came out, I had that reckoning. I saw it. I said, "I've seen this before. This looks awesome." And I saw there was an opportunity for me to actually take the skill sets that I've been working on and bet on the foundation that AI was building. I've been doing it ever since, and it's been really fun and exciting — both sharing this stuff with other people, learning more about it, and actually building my business around this.
Nate B Jones: No, that totally makes sense. Well, we have a focus for today. I want to get into some of that entrepreneurial piece, but I think we'll get into it via what launched from Anthropic. So what Claude launched — and you can add on to this — but as far as I know, the way I would put it is: yesterday Claude launched a code interpreter, which sounds boring but isn't, because it means it can create and edit Excel spreadsheets, PowerPoint slide decks, Word documents, and PDFs directly in the web interface and in the desktop app. And the devil's in the details there, because if it does it well, those four document formats are a huge part of work. Rod, how much of the work that you're doing touches Excel, PowerPoint, Word, or PDF?
Rod: Everything. Everything I do touches all of those things at various points in a week. This is awesome — from the fact that Anthropic had the foresight to see what it is that people are having a difficult time with and saying, "Let's position our product in a way to simplify that." I think that just makes for a very exciting experiment.
Live Demo: Claude's Eight-Tab Excel Spreadsheet
Nate B Jones: No, I think it is. And what I want to do as part of this is share my screen in a minute and show you some of the tests that I've been running today, and get your live reaction to dig in. Because what I did was — okay, these claims. We've heard these claims before. I think if you remember, agent mode came out, and ChatGPT was like, "Wow, you can do Excel and spreadsheets." And I did a review on that, and it was kind of forgettable. So my assumption when Claude launched this was to say they're trying to play me too, they're trying to play catch-up. That is not what I found. I found that it was actually surprisingly useful.
So this is an artificial company model. I generated a bunch of artificial data and I basically fed that data to Claude, and I fed it to the agent mode that OpenAI has, and I wanted to see specifically how it does at formulas — because formulas are one of those things that has really bedeviled spreadsheets from LLMs. And I was astonished, because you go through here and you have eight different tabs. It starts with an executive summary tab — I don't know how well you can see that, we're zooming in a little bit — where you can see the performance of the division, the annual revenue, all of this good stuff. If you click on a cell, what you'll notice is these are not hard-coded numbers. These are formulas that it's actually using. And then you say, well, where do these formulas come from? You can start to click through. This is the revenue data tab that it has created, and it actually has hard-coded these so it can reference them. These are the parameters it's using, so it's listing its assumptions. These are the scenario planning features it's using — what are its multipliers for base, optimistic, et cetera. This is the financial analysis tab where you can actually see how it breaks down the different divisions in the company: software, hardware, consulting division. It's calculating bonus calculations based on how the company did. And again, these are dynamic — you see this is equals another tab F24, so it's actually using that for calculations.
I won't pretend these are complex calculations. I don't want to oversell this. But the fact that we are talking about an eight-tab spreadsheet that is correct, that has formulas in it, that has nicely headlined, very readable headers — that is a breakthrough moment for me, Rod. I had a moment where I was like, I think that work is going to change.
Rod: Yeah, a thousand percent. I mean, I'm still copying and pasting tables from some of the LLMs or asking them to create a CSV. I'm also noticing that there's documentation.
Nate B Jones: There is. Let's go look at it. So this has a different section. It actually talks about how it defines these things — what is revenue adjustment, what is EBITDA, what are the advanced features that it implemented. And I really appreciate this because you can actually see it documented its logic. It's using a VLOOKUP here, it's using an IF statement here. Honestly, when I was doing spreadsheets, I never did documentation tasks. Absolutely not. Who has time for that? But that's the thing — it did this thing in like three minutes and it came back, and I was like, "Holy crap." When I was doing spreadsheets as a marketing analyst, this would have taken me all day.
Rod: I spent plenty of days in spreadsheets just trying to make a VLOOKUP work.
Nate B Jones: Exactly.
PowerPoint Comparison: Claude vs. ChatGPT Agent Mode
Nate B Jones: So that's cool. I think the wow factor for me was something else. And then I want to show you also — because we were talking about PowerPoints, right, and how does it do with PowerPoints? I asked it to produce a PowerPoint. Do you recall — and we may look at some agent stuff because I ran agent through the same paces, we may do that a little bit later in the call — this is what it did for PowerPoint. And again, I don't want to pretend that Figma designed this and this is the most impressive thing. Rod, folks don't know how good you are at design — you kind of hid that under a bushel in your introduction. You're good at design. You could design a better slide than this, but from an LLM one-shot perspective, this is vastly better than I've seen before.
Rod: This is actually really good. It's good, right? The spacing — it's doing the small things that you take for granted. That's a big part of the principles of design: just balance. And at first glance, this is very presentable stuff. Look at how they've chosen to center this in the boxes. That was the kind of thing that drove me nuts when I was doing PowerPoint design.
Nate B Jones: Absolutely. The spacing between the boxes themselves is about balanced. The typefaces — "executive summary" is the largest, then it gets smaller. The type hierarchy. This is really, really cool. The question I have for you is: what did you feed Claude to get this output?
Rod: Let's look at the prompts.
The Prompting Strategy: Using Perplexity to Build Prompts
Nate B Jones: What Nate video is complete without some prompts? So I'm going to go back and pull up my Claude here, and we're going to check out these prompts, because I think a lot of the key to it is — and I found this when I was digging in and comparing Claude and agent mode — let me state the principle and then we'll get into the example.
What I found is that agent mode, which is the most directly comparable example, has eyes that are bigger than its stomach. It tends to do more tabs, it tends to do a more advanced spreadsheet layout plan, it tends to get deeper into the weeds on analysis, but the execution is much more poor. So the spreadsheet doesn't have complete formulas and the layout on the PowerPoint is absolutely terrible, and you can't ultimately use it even though it was really aggressive. Whereas Claude is more conservative. It focuses a lot on tool use. It may not be as sophisticated at raw analysis, but what it delivers is very strong.
And so with that in mind, I'll show you the prompt I actually used. This is interesting — I used Perplexity to help build the prompt for this, because I found that Perplexity is really good at collecting real-time data and then combining the data into something like "make this into a prompt," where you have the self-contained financial data or whatever it is, and also the prompt, all in one thing, so that every model gets the same start.
Rod: That's really smart. And so that's an example of using an LLM to have some fun with it. And that's a great trick too for anyone that's listening — let the LLMs help you craft better prompts. The best thing is when you get into that circular rhythm. You start to really leverage the value upfront, and that's really powerful stuff.
Nate B Jones: All right, so now I'm going to go over and you can see Claude here. This is for — I did multiple problems. So I'm hopping into the prompt I used for valuing Oracle. I don't know if you noticed, but Oracle had I think a 40% pop at the open because of their reporting. And ironically they missed on top-line revenue, and I think they missed on something else too. And the reason their stock still popped was because they issued forward guidance on AI that was so strong the market reacted. So I felt like that was a wonderful timely question. I was like, what do we make of that?
Rod: Why didn't we record this yesterday before the pop so I could have invested a little bit of money into something?
Nate B Jones: Listen, right? We need to do these right when the earnings drop so we can have earnings reflections before things happen. Let's learn for next month.
So this is what Perplexity came up with. I was like, "Perplexity, set this up." So basically what Perplexity did was it said, here's the challenge — you have to value Oracle Corporation correctly. It's going to feed you data current as of September 10th. And it just makes it so easy to pull all of this. This would take me forever to pull out EBITDA margin and cloud revenue growth and this and that. So it just pulls all of that out. It pulls out a challenge: please provide a discounted cash flow valuation with sensitivity analysis. And then it goes in and says here's the current market data, here's the fiscal, here's the profitability, here's the cash generation, all of this other stuff. Is Perplexity producing this? I don't want to pretend I produced it. Perplexity did. I just asked it to do this. I had the clarity of intent. And it's a complete data package.
And what's interesting is I learned about prompting this model a little from this exercise. This prompt does not specify the output format for the model. And what Claude defaults to is not Excel. And so what you find over here on the left is Claude says, "Great, let me do the discounted cash flow valuation." It starts to kind of go through. It does all of this stuff. It's spitting out free cash flow projections, looking at terminal value, where the company should be valued, what is the sum of present value, all of that. It does all the work. It goes to the end. It has key observations. It has an investment recommendation. It says it's fundamentally overvalued by 80%. That's insight right there. The intrinsic value cannot support the current market price.
Rod: Oracle is dead, ladies and gentlemen. Time to short Oracle.
Nate B Jones: Short Oracle. Apparently that's the lesson the AI is giving us. This is not investment advice, et cetera.
And I say, "Okay, you did not do what I asked, but I didn't really clearly ask for it, so I don't fault it, because if you look in this prompt, there's nothing here that says do an Excel." And so I say, "Okay, please formulate this as an Excel." I don't ding it.
Rod: Right.
Nate B Jones: And it goes through and then creates the Excel model. And what's interesting is it gives me both a model and a user guide with documentation, and it tells me what's in the model and everything it puts in there. So I can just download it and immediately start to use it. Or I think I can actually click on it here and you can see a sneak preview. This is not really fun to look at in this view — it's more fun to look at it in Excel — but you can see it's telling the truth. It has four tabs, it has a sensitivity analysis, et cetera.
Agent Mode Fails the Same Test
Nate B Jones: So that's how we're doing. Do you want to see how agent handled the same challenge?
Rod: I'd love to.
Nate B Jones: I'm going to have to stop sharing my screen and go dig up agent here. And what's interesting is agent had exactly the same challenge — exactly the same prompt. And the response is just so different.
Rod: Isn't that interesting? Across these different LLMs, your mileage may vary every time, and some of them are just so much better at certain things. That's part of why I want this conversation, because I feel like one of the things I am taking away — one of the learnings I have as of September 10th — is that if you have PowerPoint or Excel work that you need done, you need to go to Claude for that.
Nate B Jones: You should not be messing around with anything else at this point. It just doesn't justify it.
Rod: Exactly. And you know what? This also opens up a conversation around all of those wrappers that have built their tooling around changing themes in Google Sheets or PowerPoint. This is one of those wrapper killers. You look at it — I can take that presentation and I could walk into a meeting with that. Walk right in. And that's powerful. One of the most powerful things I found in my career is that you always want to make sure that you're able to effectively communicate the things you're trying to share with people. You might have a presentation or maybe it's an Excel doc. And the funny thing is you need to rehearse that, you need to spend the time tweaking it, going through the cycles. And you're kind of doing that in the LLM as you're going through. You can actually rehearse as you're building up those presentations. And all of a sudden you don't need to stop three hours, four hours, two days before you need to go out and have that conversation, because you could rehearse up to about 30 minutes and then say, "Hey, give me this presentation because I need to go out there and I'm well rehearsed. I know exactly what I'm talking about. There are no gotchas." And that's powerful.
Nate B Jones: That's right. I love that call-out because there's that sense of iterativeness, where you are learning how to say the line, how to find the talk track as you chat with the model.
All right, we're going to see what happened over on the agent side. So you got this same prompt — go through da da da da da. So it then comes back, it thinks for longer. I want to call that out because I felt the difference — Claude would come back consistently faster than ChatGPT agent mode. And then it comes back and says, "Okay, I'm going to give you the answer." Just like Claude, it comes back with an answer in text. And so it spits out a bunch of text.
By the way, I had the text here analyzed by Perplexity for both Claude and OpenAI agent mode. And the assessment is that you get a slight extra point or two on the raw analysis for agent mode because it was a little bit more conservative as a financial analyst — specifically $57 a share, I think — and Claude had $62, and Perplexity was like, looking at the current performance, I think $57 probably makes more sense. But both of them think that Oracle is severely overvalued. So it's very much a directional thing.
Anyway, it comes back, it gives me all this, and I say, "This is not an Excel. Formulate this as an Excel." Here's the problem. Here's where things go south. You can see already there is no structure to this spreadsheet.
Rod: Yeah, this spreadsheet is dead in the water. It's unreadable. It's unusable. There's nothing you can do to pass this around.
Nate B Jones: You can't send this to someone and say, "Look at my analysis of Oracle. What do you think?" It's useless.
And this is where I got really frustrated with agent mode. I wrote up a summary of agent that I'm sure did not make new friends in San Francisco, Rod, but they were trying to say it's the best thing since sliced bread. And I kept running it through stuff like this and I was like, it doesn't deliver real workable value. I see the hype. I see the promise. But I can't get any work outputs. And that is the chasm that all of these tools are trying to solve. And I think seeing Claude do this so effectively is demonstrative of how powerful these tools are getting.
Rod: It really is. It's also changed my workflow. I'm used to the stopping point or the edge of value of these LLMs being the text and the structure, and I will go in and spend the time to say, "Okay, I need to get it to be able to play in the playground that is external to it. I'm going to have to copy and paste things." And that itself takes time. And the fact that we're seeing Claude able to actually move from within the workspace into these tools — we're talking about some game-changing stuff. And I'm really curious — I want to push this to its limit. I almost want to see it do a pivot table.
Nate B Jones: We can do a live exercise here. I think we should — in a minute here, I want to have you give me a live exercise and we'll set it up and we'll run it while we chat and then we'll go back and look at it. I think that would be really fun.
Rod: That would be a lot of fun.
Designing the Live Pivot Table Challenge
Nate B Jones: But first, I want to make you suffer just a little bit. You're a designer. You are going to, for better or worse, see what agent put together for the PowerPoint side of things. So this is that same report you looked at earlier. It's the same slide. Look how painful this is.
Rod: You know what's funny is a year and a half ago, this was really impressive.
Nate B Jones: I know. Time flies.
Rod: And then you showed me Claude, and now this just doesn't work. The bar has moved.
Nate B Jones: Look at these little footnotes — they're completely unreadable. And look, this isn't titled. It's just kind of bad.
Rod: Listen, I think what we're understanding here is that the agent is not the designer between the two. It's clearly just about the raw numbers.
Nate B Jones: And what's interesting was the conclusion that Perplexity came to. So Perplexity looked at the text outputs and basically was like, if I had to pick, I would say give your assignment to agent to do some initial analysis and then feed that to Claude to actually do the report preparation and the PowerPoint and all of that, because then you get the best of both worlds.
Rod: A hundred percent. It's funny that you bring that up too, because daisy chaining is such a powerful tool when you're working across LLMs. It's like getting the best of all worlds. So yeah, that's a great call.
Nate B Jones: All right. Well, let's have some fun. I want to go to Perplexity together and I want you to help me build a prompt in Perplexity that we can feed to Claude live and sort of have some fun with. You said you had a pivot table idea. What's your idea? What do you want to do to challenge Claude a little here?
Rod: You know what? Given the spirit of using AI, let me brainstorm some ideas as I would normally do.
Nate B Jones: Do you want to share your screen while you're doing that so we can see you in brainstorming mode?
Rod: I'm not shy enough, I think. Here we go. So I'm just going to straight up ask: help me brainstorm a few ideas around making a fun pivot table.
Nate B Jones: I love that we're using the phrase "fun pivot table." Who says that? This is why — it's 2025. LLMs have changed our entire perspective. Pivot tables are now fun.
Rod: All right, GPT-4o. Let's see what it says. Movie night. Pivot hacker. Fitness fun. Travel vibes. Recipe. These are very consumer focused.
Nate B Jones: Should we make them a little more corporate?
Rod: We could, or we could stick with this. You get the test either way, I think. What do you want to do?
Nate B Jones: Let's go with movie night. Let's make a movie night pivot table.
Building the Prompt in Perplexity
Nate B Jones: Okay. Let's throw that into Perplexity. I'll copy and paste this over to you so you can paste it in. A little team effort. We're going to have to find a way to bridge our hive mind AI.
While I'm sending this over to you via the interwebs — what are you thinking is the most unique part of this launch that made you say, "Oh, this is different. This is worthwhile"? What made you really excited about this one?
Well, I think I shared this at the top. I was primed to be bored. I was primed to not be excited. My expectation was this is not going to be that much, because I've seen this hype before. And what I found was just the ability to deliver on promises around work done became really, really compelling for me, because it meant that I could be in a spot where I felt comfortable giving a bigger piece of responsibility to AI. Which in turn — if I think about it — a lot of the question we face, that we wrestle with, is: what do we trust AI to do? How do we delegate effectively? And how much time does that give us back? How do we multiply ourselves as AI professionals? And this feels like one of those things where I can see all of these places through my day where I can start to delegate differently and I can push more to Claude because I can trust it.
Rod: That makes a lot of sense. That's one of the things I'm looking for whenever I see a launch or any of these guys — whether it's Anthropic or OpenAI — put out a new announcement. Always curious on how can it change my day-to-day, and then quickly try to get in there and tease out that value.
Nate B Jones: Exactly. Okay, I picked up the text. I'm going to go back and share my screen, and I have the text in my clipboard and we're just going to start to build a prompt live that we're going to hand to Claude and see what we get.
So the pasted text — it's movie night. And what I basically want to ask Perplexity is: I want you to take this seed of an idea and expand on it. Specifically, you need to build a prompt for another LLM to construct a pivot table spreadsheet that includes a list of 20 to 25 recent movies categorized as described below. And the pivot table needs to be very usable and user-friendly. So this is me actually prompting raw here. In addition, the prompt you construct must contain all of the data the LLM needs to construct the pivot table. Do not assume the LLM can go and get movie data — you get the movie data.
And that I found was really key, because you can't try and give it the research job and the Excel creation job in one go at this point. It will go and do research, but I find it's much more effective to have Perplexity do the research and then come back with something like a self-contained prompt, and then work from there.
And what that suggests, by the way, is that if you're working with internal company data and you're working with Claude, you're going to end up in a position where you need to be constructing a prompt very intentionally, pulling in internal company data as part of the prompt construction process, and then giving it to Claude.
Rod: Yeah, a hundred percent. And I think that's going to be a nice exercise for us as we see this first version that it spits out, because I'm already thinking about — maybe we want to just focus on say the Marvel Cinematic Universe and then we want to overlay characters, and then we want to add actresses and all this fun stuff. So I'm really excited to see what Perplexity puts out there.
Nate B Jones: Generating a complete dataset of recent movies. I cannot believe that here we are in 2025 and this is just a casual prompt that you and I are just throwing up. Look out IMDb.
One of the things that I think people don't realize — I was talking with someone earlier today who is a company leader and he uses Perplexity, he loves Perplexity, but he didn't realize that you could hit these three different options on the bottom, and he just defaults to search and loves it. And I was like, there is more to Perplexity. You can do even cooler things.
Rod: You unshackled him.
Nate B Jones: I did. I'm hoping I did. We'll see. So it gives me a bunch of recent movie titles, a bunch of viewing records data showing who watched what when, pivot table specs, expected output examples, user experience requirements. This is very complete. This is very thorough. Do we want to just run it and see what happens?
Rod: Absolutely.
Claude Builds the Movie Night Pivot Table Live
Nate B Jones: All right. We're going to go run it and see what happens. I'm excited. I'm going to stop sharing and paste it into Claude and we're going to see what we get.
All right. So we're going to copy it. And I'm glad I told it to be self-contained because there's no way I would be writing all of that stuff out. All right, so I'm pasting it in, and I actually want to do this live. So once I get it queued up, I'm going to share my screen again and we're just going to see how it goes.
Okay, here we are. And literally it pasted in, and you can open it up and look at it. You can see this is the instruction — it does specifically say "create a spreadsheet," so I'm hoping I don't have to remind it to create the spreadsheet this time. It has all the data we looked at. In addition, it dumped in a bunch of URLs because Perplexity just can't help itself. That makes the prompt slightly dirty, but I'm just leaving it in for now, and we're going to see what it does.
Rod: I love how Perplexity loves citation.
Nate B Jones: It really does. And so it's starting to go through. It's going to create — so it's first analyzing it. So it's got the Python script up here. You can actually see it starting to code it in, which I find just addictive to me. I love this.
Rod: Yeah, the world is a different place now. It's just wild. It's so funny that there was a time where doing something like this was extremely time-intensive. And I say "a time" as if that wasn't just a couple of years ago. It just feels so far away. We are such creatures of convenience that the moment you put things like this in front of us, we were like, "Give me. This is the new floor."
Nate B Jones: And so it's just going through, and as far as I know it hasn't even touched the spreadsheet yet. All it did was use Python to figure out what it was going to do, and then it's decided, okay, now we're going to build a pivot table and actually start to bring in an Excel file with advanced features. So we'll see how it does on that.
Rod: I really love the way it thinks. I love the way it does the thinking out loud, and the way that it's leveraging Python to give it a pathway so that it can then use that. And I think that's something I'm noticing across the different LLMs. The engineers are getting smarter about this — they know that there are certain languages that just lend themselves well to going from human language to computer language, and Python is just one of those languages. So if you can get into Python, all of a sudden shifting into other languages becomes a lot easier because you're getting that structure built in. Really, really cool stuff when you're actually paying attention to what's going on there. You're not just pure vibing, which I love to do too. A good vibe is great.
Nate B Jones: I think that when I'm able to pay attention, I notice that Claude is much better about this than OpenAI. Like the tool record here is really clean. I can see exactly what it did all the way through. And now — it loves documentation, so it's creating me a readme. I didn't even ask it to create me a readme. This is so great. "How to use the spreadsheet." Thank you, Claude. Sometimes Claude is a little bit stuck up, but I love it so much. I don't mind it.
Rod: Can you laminate it? Can you collate it and mail it to me too, please, Claude?
Nate B Jones: You remember Calvin and Hobbes? Claude has the personality of Susie Derkins to me. It's that kind of a model.
Rod: That's a great callback. Yeah, I love Calvin and Hobbes. It's so good. My kids are addicted to it.
Nate B Jones: Okay. Oh, wow. It overachieved. It has a complete pivot table and apparently an enhanced analytics version. What? Amazing. Okay, we have to see what's in here. So I'm going to download both of these and pull down the readme. I actually want to open these in Excel because this is just a crappy way to look at it. So give me a second. We're going to look at this in Excel and see what's in the box.
Rod: Excited. While you're pulling that up, I'm also excited to see how Microsoft and these other companies respond to this, right? The amount of dynamic behavior that they're going to have to start introducing to their products. It's just opening so many doors.
Examining the Results: Heat Maps, Spark Lines, and Over-Delivery
Nate B Jones: So this is the enhanced version. It did color coding. Heat coloring. Yeah, it heat maps. I'm kind of impressed. I didn't ask it to heat map. It volunteered. It did pattern analysis. Look at this. I'm kind of blown away. I didn't know it could do this. Rod, this is also me reacting in real time. Are those — do you zoom in? They're like spark lines or something. Little smart graph line thingies. Whatever you call these things.
Rod: Spark lines.
Nate B Jones: I don't even know. Like little bar graphs or something. Yeah, they're little bar graphs right in the cell. And then it has this cute little — okay, take away a little credit because the genre is on top of the bar, but whatever. You get friend statistics so you can see what these different friends like — unique movies, diversity, et cetera. You get your top 10 movies ranked. So this is the Susie Derkins overachievement version. And then it has all the raw data. And the raw data — that's what we love to see.
Rod: That's so good.
Nate B Jones: Yeah. So you can tell I'm not lying — that's the implication. And so there it is. That's the enhanced version. I'm going to go ahead and share with you what it called the basic version, which I thought was still perfectly usable. And you can kind of see what it thinks is special, which is its own little piece of insight to grab. So yeah, this is it. This is regular movie night without the heat mapping. So it's exactly the same table. You see the 12 is there, and all you have is a nice clean traditional finance table with blue and white rows, data perfectly understandable. But you know, they segment out each of the viewers — that was not in the enhanced version. So they actually have data on Alex, Jordan, Casey, Morgan, and Riley, and you can see how they go by day, what their genres are.
And so in a sense, one of the things I am learning from this is that the more interesting you make the prompt for Claude, the more likely Claude is to give you some permutations. And what I might do in a follow-up here is basically say, "Claude, I loved the heat mapping on the pivot table. Can you please preserve the heat mapping, but also give me the viewing pattern pivots and do a modification?" That'd be really cool.
Rod: Actually, open up the pivot table. Can we pull in — can we go into the data?
Nate B Jones: Where's data? Up here. Let me take a look. It's been a long time since I've been inside Excel. I will tell you that. Are you looking for something specific?
Rod: I'm looking for the pivot. There's a table that you could actually see how it creates the cross-relationships.
Nate B Jones: Is it in analysis tools? No, it's not. What if analysis, text to columns, flash fill, queries and connections. That's it, isn't it? No, that's not it. We'll have to find that later. I don't think we're finding it right now.
Rod: That looks really cool. This is awesome.
Nate B Jones: Yeah. No, it's super fun and I'm learning a lot too as we chat.
Takeaways and Closing Reflections
Nate B Jones: So there you go. That is what Claude has launched. Do you have any takeaways from this? What are you taking away from this conversation?
Rod: For me, what I'm realizing is that the idea of using these tools is no longer relegated to just me as a person. The amount of decisions that I need to make around getting something valuable out of the outputs that I'm getting can really, really start diving into the more interesting bits. One of the awesome things that we've seen over the past couple of years with these LLMs is that we all get through the first 20, 30, 40%. And most of us, by the time we're really proficient at what we do, the 80% is pretty much — you could phone it in. But it's that last 20% that makes things go from good to great. And the fact that we could spend more time focusing on that 20% — and this is a perfect example of that — makes for a really interesting future in both the professional space and just the communication aspect. This is really, really cool stuff.
Nate B Jones: So I want to give you a philosopher quote that I promise is relevant, that I am taking away as I look at this. This is from Alfred North Whitehead, who famously said, "Civilization advances by extending the number of important operations which we can perform without thinking of them."
Rod: That's actually a really good one right there.
Nate B Jones: And I feel like Claude extended civilization, because there are important operations where I didn't tell it to put the heat map in. I didn't tell it to center the text on the PowerPoint slides. It just did it, and I didn't have to think about it.
Rod: That's awesome. Yeah, I like that. I also like that positioning. Anthropic might come knocking, Nate. At this point, you're saying that Anthropic by extension is extending civilization. Just fantastic.
Nate B Jones: Well, if Dario calls, I'll pick up the phone.
Rod: Well, thank you for having me. Thank you for chatting a little bit. Thank you for making time. I know this was a bit of a last-minute thing. We're having fun.
Nate B Jones: First of many. Loved it. Looking forward to doing another. Have a good one, Rod.
Rod: Cheers.