8 Jev Use Cases That Feel Like Cheating transcript

Jev is everywhere and it is the most exciting AI that I have seen in a long time. It really does feel like the early days of chat GPT where we had this thing that was so powerful and so raw and people are trying to figure out how to get the most out of it. And so I'm going to try to explain today not only what is Jev, but also how to use it, how to think about it, what it's good at, what it's not good at, and then I have a bunch of demos of things that people have already built with Jev.

All right, so first, what is Jev? Jev came out just a few days ago, and it is a new type of artificial intelligence that is basically just a really awesome decision maker. But it's more than that. What really sets it apart is that it can make thousands, tens of thousands, hundreds of thousands of decisions incredibly quickly and incredibly inexpensively. And if all of this still sounds very handwavy, stay around because I'm going to give you some actual concrete details about what this thing actually can do.

All right, so here is a very basic example of how Jev works. You have questions and you're structuring this as JSON. So the instructions or basically the prompt is is hot dog a sandwich? Okay, you give it criteria basically how to think about the question how to know whether yes it is or no it's not a sandwich and so true a sandwich is a food dish where a filling such as meat, cheese, vegetables or spread is placed between structural starch.

False. the food has no bread enclosing a filling or uses only a single slice of bread or uses a non-bread wrapper such as a tortilla, wafer or cookie. Then up here we can actually go further. What is food? And then we have the definition. And so with all of this, if we run it, is hot dog a sandwich? 73% true. So it would resolve as true here. And so Jev is really good when you can describe the type of decision you want to make, when you can provide really clear context and structured questions.

If you check all of those boxes, Jev can be incredible for your use case. But there are things that it's not good at. For example, it's not a text generation model. You're not going to ask it an unstructured question expecting an answer. For example, you're not going to ask it for advice about how to talk to your boss about something. It's just not going to work that way. And you're not going to ask it to write code for you either.

It's not that good at writing code. But something it could be really cool for is if you have a large library of front-end UI assets and you want to create a front end dynamically. It can actually choose from that UI library and put together an entire web page for you. But it just requires that you have a bunch of buttons that it can select from, a bunch of input fields that it can select from, fonts, and everything else.

And then it can put together an incredible website for you in less than a second. So if you put Jev up against Fable in a game of chess, Fable will win. However, if you put Jev up against Fable in a game of chess with a time limit, Jev will most likely win just because Fable takes so long to process each step. and Jev can make a decision nearly instantly. And so most likely even though Jev wouldn't have won the game with no time limit because the time limit will run out on Fable, Jev will win.

So here's the first example. It is quite a good slop detector. So if you think that a website or some kind of creative writing has been written by AI, you can actually have Jev check it for you. And there's this website madewithjv.com that has a bunch of different examples and demos of really cool things that have been built with Jev. So that's where I'm pulling this from. So we checked anthropic.com and it is 26% AI slop, mostly handmade, mostly humans, but there is some slop there.

And that's to be expected. They are obviously using the cloud models to write some of their content to say the very least. and it actually breaks down specifically what it's looking for and why it determined that a part of the website was slot. This is another incredible use case for Jev. I really don't understand how this didn't go more viral than it did. It only got 36,000 views, but this is by Kitsy and it's Unclutter, a smart ad plus slot blocker with Jeb.

It is nearly instant and as you can see it works incredibly well. So, it auto cleans websites as soon as it loads from all ads, cookie banners, upsells, and BS dialogues. You plug in your own key, and it again is nearly free. And I should actually talk about the pricing for a second. Jev is 4.2 cents per million input tokens and free for output tokens. Unlimited output tokens, absolutely free. And so what's actually happening when you load a website is it's sending all of the elements to Jev with a description of how to determine whether something is spammy or annoying and Jev will make that decision and then it will remove all of those elements.

So very simple, very cool. The next one is actually the sponsor of this video. It's Zapier. I've been working with them for a while. They actually added Jev into the Zap platform. So if you're building any type of automations, you can use Jev right in the workflow to make decisions for you. So for example, let's create a new workflow. And when you come down here to actions, if you just search for Jev, there it is. Type safe Jev.

Now you can use that as an app in Zapier. So, so you could take any of the over 9,000 different apps that they already integrate with and connect them in different ways and have Jev make decisions between those connections. And so, for example, I can say, okay, every time I get a new calendar invite, Jev decide whether I should accept it or not. And I'll give it a bunch of criteria as to how it decides that. But now, it'll automatically either accept or decline it.

And obviously this is a very simple use case, but you can get as sophisticated with all of the over 9,000 different apps as Zapier has to offer. This is one of the coolest use cases that I've seen for Jev and Zap Connect 2026, which is their annual conference, just happened. Check out my link down below to see what's possible with NextG Zaps and Jev. All right, here's another example of what Jev can do really well. Typically, email clients are sorting in chronological order, meaning the last emails coming in are the ones that you see.

And I've built things in Grockbot to help me prioritize the emails I should respond to. But Jev can do that basically instantly. Thousands of emails can be sorted by what it believes is the priority in response. And so that's what this guy built right here. And 100 emails getting sorted in under a half of a second. Here's another great use case actually valuable. You know how you can do find in page on browsers? So if I type like that and I type find, it'll highlight the words.

I use this all the time. Super useful. But it is keyword matchingbased. So if the exact thing you type isn't on the page, it's not going to show up. Now this product manager from Google put together a fuzzy search version of it. And typically it would be quite slow, but now it's extremely fast. You just type in whatever you want to find on the page. It doesn't have to be exact. You just have to type in kind of what you're looking for and it will highlight on the page exactly where to find it.

And of course, it's open source. It's free. You can bring your own key. It's just awesome. Okay. And you remember at the beginning of the video how I said it's not good at writing code, but if you have a preconfigured library of UI elements, it can actually build a web page very quickly for you dynamically. This is what I mean. On the left, we have a page rendering and on the right we have Jev doing it in real time dynamically.

And look how fast that is. It basically decides, okay, based on what this user is looking for, what should the website look like? Do they need a form? What information do they need? Do they need to sign in? Which is what we're seeing here. And it's all extremely fast. But again, it's not building it from scratch. It's just choosing from a bunch of different UI elements and putting them all together in the right way. Okay.

And this next one we might be using pretty soon. This is an example of Jev clipping a 90 plus minute video into multiple clips based on whatever you type in in under 2 seconds. So, here's an example. This is called Clipfast. You type Y RL works. It is scanning the entire video, probably the transcript, and finding all of the right clips and instantly clipping them. Then, one click, you can just download it so easily.

I mean, this is the type of use case that speed really unlocks. Okay, I have two more left. And not all of them have to be super high value. Sometimes they can just be awesome and cool to watch. Here's an example. So, here is Jev understanding color palettes. You literally can type anything and it will put together what it believes the color palette is. So, here's 80s disco 80s pop retro. Here's browser, like what a browser looks like.

And here's what Bowser looks like. Here's Mario. Here's Luigi. Wario. So, it knows what colors are associated with each of these world knowledge entities. Here's the blue screen of death, the matrix, the beach. So, really cool. Just an awesome demo. And then last, we have a pile of emojis at the bottom. And you can just start typing something, anything, and it will decide which emojis are most associated with what you're typing.

So, here I need to lose weight. Of course, it has healthy food emojis. Starting a band. There it is. We have instruments. So, kind of fun, quirky, simple, but I think it's really cool. So, go check out Jev. Start using it. It is nearly free. They give you $5 in credits to start out, and it is just so much fun to play around with to see what's possible. So, hopefully these examples gave you some inspiration about building something yourself.

And I made another video going over in depth what Jev really can do and what it is. and check that out right