How to Setup Jev AI + Claude Code
AI News Today | Julian Goldie Podcast
September 22, 2026
How to Setup Jev AI + Claude Code
Speakers Julian Goldie
Julian Goldie (0:00)
Today, I want to show you how to use one of the most powerful AI models, one of the most revolutionary AI models that just dropped, which is Jev AI.
And I'm going to show you how to use it with the Claw and how easy it is to connect both of them together. Now, there's a takeaway and there's a freeway. I'm going to show you both. And then you can decide which one you want to use. So, for example, if you want to use Jev with the paid method, you can actually go over to Open Router and you're going to get Jev Latest like you can see right here. Then you're going to click on Create an API Key. And from here, you can just create an API Key. If you're not sure how to find Jev on Open Router, you literally just go in the search box, type it in like so, get Jev Latest, create API Key like so. And what you can also do is just create a key like this. We'll call it Jev. Once you've got that, you can just copy the API Key like so. We're going to go back over to Claude now, and then we are going to click on New, paste in the API Key, or you can put it in the EMV if you want to keep it more secret from Claude, which actually I'd recommend, to be fair. And then you're going to grab the details of Jev and how to use it. So you can see the documentation here from OpenRooter, and then you just go back into Claude like so, and you paste that into OpenRooter as well. So it's got a lot of documentation. It's got the API Key from me. And then we're just going to say, OK, here's my Jev API Key from OpenRooter. Based on what I work on each day, give me some ideas of the best use cases for making the most of Jev and some practical, actionable ways to actually use it. And the reason that I recommend that is, number one, not only do you connect it quickly to Claude, which is super easy, right? You've given it the details, the documentation. And also Claude has web research, so it can easily understand what Jev does and how it works and everything else. But also because it understands you, right? Claude has access to your memories and everything that you've worked on recently and all of your recent projects. And because of that, it can actually work through what you've done lately and then give you some ideas on how you can actually implement it based on your day to day workloads, right? So, for example, if we have a look through this list, these are all use cases that are personalized to me. So if we go down this list, it's like, okay, well, you could create a machine that basically does the internal links across your website. And here's how you would use it. Or, for example, we could use it with Google Search Console, and we could basically give it some choices. It can score the keywords. It can make decisions for us. And that's the whole point with Jev. It just makes decisions very, very quickly. And that's why it's cheaper and faster to use than something like Claude. This is the other thing about using Jev with Claude. It's not only is it pretty powerful, but also it can make Claude itself a lot cheaper. So instead of giving decisions, for example, like what's the search intent of this keyword, or what are people looking for with this SEO keyword?
Well, with Fable 5.1, it will be very, very slow to understand that. And sometimes it would actually get it wrong. With Jev, it would actually look through the choices and be like, right, okay, these leads are good. These leads are bad, right? And it can easily sort through like leads, or it can quality control stuff. It can check through keywords, but it's just very quick at making decisions. And that's why you can use it with something like this as well. Here's another example. So we could actually look through our agentic operating system, which we have over here. And we have a system with Hermes Oracle, where it pulls in the latest news, right? So it pulls in the most trending news right now. And it basically every 24 hours updates with what's trending and what's the most useful, like so.
Now with this system, if we were using Jev with Hermes Oracle, we could actually run Jev through the latest news and be like, is this worth creating content about or not? And this would use something called Null, which is one of the decision making processes that Jev can use. And then it can have a choice. Is it worth it or not? And then only items at 0.8 confidence score. So only if they have a very high confidence score, do we create content around them. Otherwise, they would just be ignored. Or another example, you could be like answering comments across social media. And it could be like, okay, it's going to score them on urgency, or it's going to use a choice system where it's answering a question or sharing a win, or it's answering complaints, or some stuff is just worth ignoring. So you can see how it's basically giving me very personalized ideas. So step number one is you get the API key from OpenRouter. I've shown you how to create it. I've shown you how to get the documentation. Then you're going to give that to Claude. From Claude, you're going to ask it, okay, what are the best ways we can use this in my business right now? And then I would just pick one. I'd look through the list of like, for example, we've got 10 options here. I'd look through the list and be like, okay, this is worth doing, this is a waste of time, and you just pick one, right? You just go with one, which is pretty cool. I'll show you something that we built that was really cool. And then I'll show you how to get access for free as well to this. So you can see, for example, here, we have a system where it can basically watch me live, as you can see right here. So this is me live, right?
3 more minutes of transcript below
Thousands of transcripts fetched by people building searchable podcast archives
Fetch the whole transcript
The demo key returns a sample episode in full, no card needed:
curl -H "x-api-key: pt_demo" \
https://spoken.md/transcripts/1000651996090Markdown with the speakers named, for your notes, your knowledge base, or anything that makes HTTP calls.
From $0.10 per transcript. No subscription. Credits never expire. Prices exclude VAT, added at checkout for EU customers. Not what you expected? Email us within 14 days with 20 or fewer credits used and we refund the pack in full.
Using your own key:
curl -H "x-api-key: YOUR_KEY" \
https://spoken.md/transcripts/1000791143244