Nate Herk gave Claude Code a bad coffee shop website and prompted it to 'make it better' 100 times in a row, comparing Sonnet and Opus results and milestones.
Key Takeaways
- Repeated 'make it better' prompting yields diminishing returns and tends to bloat code and features.
- Models need explicit constraints and human taste to push quality beyond a capped level of improvement.
- Sonnet and Opus both produced meaningful early milestones, then mostly incremental or overengineered changes.
- Some later changes broke existing functionality, so supervision matters when running long autonomous iterations.
What the video covers
- Nate Herk tested Claude Code by repeatedly prompting it to improve a deliberately bad coffee shop website, 100 rounds in a row.
- Each round ran in a clean session with no skills, CLAUDE.md, plugins, or memory, relying only on the model's knowledge and built-in tools.
- The prompt required looking at the site, innovating outside the box, making one bold change, and shipping a change every round, since models otherwise stopped around round 20-25.
- The experiment tested Claude Sonnet and Opus at medium effort, with Sonnet's run estimated at about $29.93 in API billing.
- Sonnet's milestones included a first image, first JavaScript, mobile readiness, a sticky order bar, an AI-generated background video, and a working online order form around round 32.
- After round 32 Sonnet had no major milestones through round 100, while code grew steadily and linearly, raising doubts about whether quality improved.
- Sonnet's round 100 version added features like testimonials, scrollable order panels, popups, and urgency messages, but Herk judged some changes to be fluff.
- Opus was also tested and tended to overengineer, adding a caffeine calculator and a flavor-matching quiz, and some sections broke as a result.
- Neither model performed web searches for design inspiration, which Herk had expected them to do.
- Herk's conclusion is that models hit a ceiling on vague goals and that human taste and guidance are needed to get from roughly 80 to the high 90s.
Full Transcript — Download SRT & Markdown
Speaker A
Okay, so this one's a little bit different because today what I did is I gave Claude Code this website which is a horrible coffee shop website and I literally just told it make it better a 100 times in a row. So what I want to
Speaker A
talk about today is basically what happened. I did this with sonet 5.5 on medium effort and I did it with obus 5.5 on medium effort and I just said make it better 100 times. So I'm going to show
Speaker A
you guys what happened. So as you guys know this is the original site. I had to make something completely awful and ugly for them to start with and then this is the exact prompt that was shot off to
Speaker A
them 100 times in a row. So it says, "Look at the site first, then innovate.
Speaker A
Think about the user experience and think outside the box. Make one bold change that improves conversions or the look and feel, not a small tweak. Use any tool you have. You must ship a change every time." And the reason why I
Speaker A
had to be so specific with this prompting is because when I tried this before, it would just stop at like 20 or 25 and be like, "Hey, nothing else can be done." And so I had to basically force it to make a change every time.
Speaker A
Now, the reason I wanted to do this is because lately I've been having so much success building different AI pipelines when I've been using inspiration from other people or other proven methods because I think AI is really, really bad
Speaker A
at thinking outside the box. And a lot of us would love to just be like, "Hey, make this prettier." Or, "Hey, make this, you know, smoother." and and just kind of give these big vague goals which are good but I think you hit a cap at
Speaker A
maybe let's just call it like 80 out of 100 and so I think to get to more of the 90s you know high 90s 100 out of 100 you have to be the one that's applying the taste and helping innovate so that's why
Speaker A
I wanted to try this experiment now another important thing to know before we hop into the results is that it ran in a clean room every time so fresh session no skills no cloudmd no plugins no memory each model had basically just
Speaker A
its own knowledge and any built-in tools so it looked at the site like I head. It thought about it and then it decided, okay, here's one thing that we can do to think outside the box to improve the
Speaker A
site and then let's ship it. So, here's the little local host that it spun up.
Speaker A
You can see that I've tried different rounds here and we're going to be looking at Sonnet on Medium and Opus on Medium. So, let's start with Sonnet. You can see on this left, it's going to show you every single round, how many tools
Speaker A
were called, how many lines are in total the site. But really, what I want to be looking at are some of the milestones.
Speaker A
First of all, what you can see is that Sonnet ran and this would have cost us $29.93 via API billing. Sonnet also created 11 images and two videos in this site. But anyways, I'm just going to go ahead and
Speaker A
switch this over to the milestone view. And that's going to show us when major things happened. So, in round one, we obviously had a major milestone. There was the biggest jump here because we had it was the first time it looked at its
Speaker A
own site. It had the first ad generated image, which is this background right here. It was our first JavaScript and it was our first mobile ready. So, here is V1 of the site. Been there, coffee, being their coffee. We have an image in
Speaker A
the background. And I can actually open this up in a new tab so we can look at it a little bit better. We have a very simple looking punch card, very boring font. All this stuff is just very, very
Speaker A
plain. Nothing too crazy. If we go to round two, we had our animation come in.
Speaker A
So, this is where we had a first animation. Let's see where that is. I actually think it it was just this when we have our order and we scroll, this little order bar sticks. So, if we started filling things in and then we
Speaker A
scroll down, we would get that to be sort of like stuck here. And that's not bad, right? That's a nice little innovation. We didn't have another major milestone until round seven. And you can see here the lines of code went from 253
Speaker A
up to 419. And this now says the milestone was first working image on the page. So, it looks like now in the hero section, we have an image up here. Um, and maybe this wasn't what it's referring to, but we can choose things
Speaker A
for the menu. Where is the working image? Maybe it's these things. when you add them into the menu, you can actually see a picture. I think this was probably the milestone that it's referring to.
Speaker A
But anyways, you can see it's just slowly making tiny tiny little changes. It's definitely getting better, but it's definitely not great still. Now, we have another milestone at round 24. 778 lines. We have a first AI generated video, which looks like this background
Speaker A
right here that we see where we now see this little animation coming through in the background. We also see that this has been improved. We can also see that this has been improved. We got a nice little scroll animation with a, you
Speaker A
know, a static image in the background, which I think is a nice touch for some depth. We also have this has been improved a little bit. The order now looks more like a receipt or like an actual ticket rather than just a little
Speaker A
weird animation box. And yeah, we can see that we're definitely getting things added. We can see that our punch card is showing up. And yeah, it still keeps the same branding and everything, but we're adding little things. And actually the
Speaker A
final milestone here was round 32 where if I open this up it says this was the first form. So we have an actual form now it looks like where we can order online maybe and then we can order ahead
Speaker A
and go pick it up. So we can maybe schedule our order here. This takes us up here where we can add things to the order. Let's see if I add it. Then we can actually place that and boom we
Speaker A
could then go pick it up and we have our food slashdrink. So functionally it has increased the user experience a little bit. It hasn't done anything too crazy as far as the site, but you can see that each round we have had some sort of
Speaker A
small innovation. Now if I turn this off milestone only view, obviously there's tons of other rounds. It didn't actually say anything was big enough to be a milestone from rounds 32 all the way to round 100. Now what's interesting is
Speaker A
right here you can see API value per round. You can see lines of code and what happens is the API value is like kind of going up and down and the lines of code every single round is steadily
Speaker A
increasing kind of linearly and what that tells us is like it's adding a bunch of code and we don't know if things are actually getting that much better. So here is the round 100 version from sonnet. We can see we had a little
Speaker A
animation there with ordering ahead and walking in line. I'll refresh in case you missed that. We have a nice little floating bar that kind of moves up and down. That was probably one round of innovation. It's got a little glow to
Speaker A
it. Now we can see the full menu. We have some hero stats up here. We can switch between help me pick up, browse, or just typing in your order. We also have a little popup now. So, there are
Speaker A
definitely other things. We even have like only two left, baked at 11:00 a.m. So, it's definitely getting better. It also decided to make this little order thing um scrollable, too. So, I like that. We can see we even have now some
Speaker A
testimonials in here. We can order their usual orders. We have some little leaves falling down here, the maple cinnamon latte, and all of these other things.
Speaker A
We've got a scroll bar or a scroll animation bar of actual products in here. So, more videos, more videos. We can see how long the line is. We can see the typical busy hours, all this kind of stuff. So, the question really is like
Speaker A
at what point is this too much, you know, like what round would we have stopped and been like, "Okay, you know what? Round 65, this is good. Anything else that you keep adding is just going to be extra fluff. We're not going to
Speaker A
need it." Interesting. It actually looks like it took something away. So here is round 65. We have like this little navbar on the right. But what'll be interesting is if we went back to round 100. Did we have that? I think we don't.
Speaker A
So it ended up actually taking away a feature that it didn't like which is pretty interesting as well. But I think at a certain point it just like starts overengineering because obviously what I told it to do is you have to ship
Speaker A
something. So it starts forcing changes. Like here you can see this is round 74.
Speaker A
Look at this font. If we go to round 100, you can see that that font is different. So at one point I was like, you know what? I'm just going to change the font. So, very, very interesting.
Speaker A
But let's pivot over now to Opus. Real quick, guys, I've got this completely free website design skill that I'm giving away. I use this for every single website that I build, including our AI Automation Society site, which I think
Speaker A
is pretty slick, and I absolutely love this website. It's called ScrollCraft, and it understands things like scroll driven animations and layering, but also it has things in there like taste, typography, spacing, depth. It's just a really good website design skill in
Speaker A
general, and it's completely free. So, the link for this will be down in the description, but let's get back to the video. All right, so same exact thing.
Speaker A
We're going to start with the milestones only, and then we'll look at some of the other ones, but this one cost $137, 54 AI images, and three AI videos. So, the first round, V1, first time looking at its own work. First AI generated
Speaker A
image, first animation, mobile ready. So, it definitely made more progress on round one than Sonnet did. We have a picture here. We've got the classic anthropic font. We have the menu, which we can select some items, and we're
Speaker A
already getting a bit of a popup down there. We can text our order. Once again, it decided to do a punch card.
Speaker A
Well, it probably was on the original site. We've got the hours. Okay, cool. So, very bland, very boring, nothing too crazy. Round seven, we had our first AI generated video, which looks like this background of the hero section, and it
Speaker A
starts to look a little bit more like that sonnet version. We have a very dark theme down here. Two days off, the roster. We can choose different flavors.
Speaker A
We have a bit of a menu now with some different AI generated images of these items, punch card, and visit us. So, nothing too crazy, just the video. Now, round 15, we had our first form. So, now we have the ability to actually fill
Speaker A
something out and we can start to like send in this order. We can also do a QR code, which is pretty cool. Actually, let me see what happens when I fill out that QR code. Oh, interesting. So, when
Speaker A
I scan the QR code, it fills out a text. I don't know if you guys can see this.
Speaker A
It puts my order that I put in the actual site and it sends it to the number. So, I could literally just text this order punch card one of nine after this order. Wow, that's actually not bad. That's pretty cool. So, that was
Speaker A
our form. Nothing else is too crazy. It's starting to play with colors a little bit differently on different sections. It's also showing us that this is roasted 20 ft from your cup. So, we get a cool little animation there, but
Speaker A
besides that, nothing too crazy. And then the last milestone here was round 32 where we have our first canvas or webgl. And I actually don't exactly know what that means, so I'm just going to call Glido real quick. Hey Glido, what
Speaker A
does a canvas or WebGL mean inside of a website? We should see up here, Glido will shoot that off to the web. Canvas is an HTML element that lets scripts draw graphics like shapes, images, or animations directly on a page. WebGL is
Speaker A
a JavaScript API that uses the browser's graphics hardware to render 3D graphics inside the canvas, enabling complex visuals and games. Okay, interesting.
Speaker A
So, let's see if we can find that. I wonder if that's this. Maybe it's this sort of animation where we can say, you know, what are we looking for as far as something easy or waking me up and sweet
Speaker A
and cozy? So, wherever you decide to put this, it will basically say, "Hey, this is the drink that's most close to what you're looking for." So, that's actually pretty cool. I think that was one of the innovation times where I actually would
Speaker A
have kept that change. But, I feel like the majority of the times when I'm asking Opus to just innovate, I'm not going to actually accept that change cuz I maybe think it's overkill. So, I like this little whiteboard. I think that
Speaker A
there's too much text on here. Honestly, this is supposed to be like people's orders, which I think is kind of cool.
Speaker A
But anyways, we can start to see different timing and when we'd be able to get it. That's pretty cool. good for the user experience. Also, we can like reserve seats, which is pretty interesting. Not too bad. Where to pick
Speaker A
it up from and how long it'll take us. So, anyways, I really like this text at the bottom, too. So, now we're at the point where like I might be okay with this. I might stop and be like, this is
Speaker A
already a lot, but we still have 70 rounds to go, 68 rounds to go. So, let's go out of this view and let's first look at the very finished version. So, round 100. If I open this up in a new tab,
Speaker A
this is going to get overwhelming. You know, Opus was forcing iteration here. And there's a lot going on. There's leaves. The background has changed.
Speaker A
There's so much text. So, this is the point where it's like, yeah, this is AI slop, right? Like, look at this right bar. Look how much we have to go through. Look how much we have to go through here. We can carve a pumpkin if
Speaker A
we choose a small knife or a big knife. You've got to be kidding me. And this doesn't even work properly. Well, I guess it kind of does, but not really.
Speaker A
Like, you have to start at the bottom. So, that's kind of buggy. It didn't really get tested very well. But as you can see, we're just at this point now where this is definitely overkill. We don't need all this info because people
Speaker A
probably just want to grab their coffee. And this se this section doesn't even work anymore either. So, it ended up breaking some of its own stuff. It's also looking at like caffeine, where it will be in your system, how you'll be
Speaker A
able to sleep. You can also toggle this on if you want to go to Spanish, which didn't even switch this to Spanish. So, it ended up trying to force so many changes that it broke the site and it
Speaker A
just killed the user experience. Like, this is just obviously way too much here and we're not even halfway through the site. We can now have music. We can choose so many different things in here.
Speaker A
So many different Like I like this review board actually. I like the vibe of this, but it's just too much, right?
Speaker A
Like it's gonna get lost in here. I think you guys get the point, right?
Speaker A
Like it's very interesting just to see if you force Opus to keep working, it is going to just overengineer things. It's going to overkill them and you really need to be there to give some guidance.
Speaker A
So hopefully this was interesting to you guys as well. If we look at the final stats, Opus ran for eight and a half hours. Sonnet ran for almost two. Opus spent about 137 bucks. Sonnet says Sonnet spent about 22. Obviously, if
Speaker A
this was API billing, I did this on my subscription, but obus 5.5 ended up having 21,000 lines of code in this site and Sonnet had less than 3,000. We have a difference in AI images and videos.
Speaker A
And the first video for each of these models, Opus did it in round seven and Sonnet did it in round 24. Now, once again, I think this is really interesting because it didn't utilize any skills. It didn't look at any
Speaker A
reference sites. It just tried to do its own stuff. But what really really shocked me, which I was honestly kind of upset about, is that it didn't use the internet at all. I thought I put that here as a bullet point. I guess I
Speaker A
didn't. But it didn't do any web searching. I thought for sure that it would be web searching websites and maybe looking for components at something like 21st. Or looking up websites that are winning awards and try to steal inspiration from that. But it
Speaker A
didn't do web search. None of these models or I guess not none of them.
Speaker A
Neither of them did web search which really really shocked me especially when I kept saying innovate innovate think outside the box. I thought that it was going to do some research to pull inspiration from but it didn't. So the
Speaker A
model sets the ceiling and the prompt sets the direction. It was obviously the same prompt here but the different models came out with very different results. Innovate without a vision turns into adding. They started adding stuff and forcing stuff in when we didn't
Speaker A
actually need it. It also really really wanted to improve conversions. So it tried to upsell everything. It tried to force ordering. It tried to make it so so seamless but it was too much. it was like a hard sell rather than more of
Speaker A
like a soft close. So, very, very interesting. Like I said, this isn't going to change a ton about the way that I work, but I just wanted to know and I hope that this was interesting to you guys as well. So, if you enjoyed the
Speaker A
video or you learned something new, please give it a like. It helps me out a ton. And as always, I appreciate you guys making it to the end of the video and I'll see you all in the next one.
Speaker A
Thanks everyone.
Topics:Claude CodeSonnetOpusAI experimentprompt engineeringweb designiterative improvementAI automationcoffee shop websiteLLM overengineering











