Full Jev Video Tutorial

The post itself gives you a chapter list and nothing else: intro, Jev explained, API setup, then demos of a voice browser, AI memory, and a YouTube predictor. Every prompt, code listing, and cost figure lives inside the video, and the post documents none of it.

You cannot tell from the post. It names an API setup chapter at 4:06 and no more, with no SDK version, endpoint, or model identifier anywhere in the text. Treat it as launch-week material and check the current TypeSafe documentation for the real setup.

The post lists them by name only, with no repository links. One of them corresponds to a separate project the author maintains, jev-voice-browser, which is the single lead on whether any of this is reusable. The memory and predictor demos are undocumented in the post.

Nothing in the post suggests it does. The chapter list runs from intro through API setup to three demos and stops, with no chapter on spend, backoff, or failure handling, so the post gives you no way to judge whether the video spends time there.

On YouTube. The author posts a follow-up saying it is also on YouTube if you want to watch on your TV, under a title calling Jev the new AI model breaking the internet. The chapters and demos are the same; the YouTube copy is just the tutorial in a longer-form player.

Moritz Kremb, who describes himself as someone who tinkers with AI and shares what he learns, and who maintains jev-voice-browser, the project behind the voice-controlled browser demo. He also points to a free resource community for mastering AI tools, so the tutorial sits inside a broader teaching operation.

Six chapters: an intro, about three and a half minutes on what Jev is, API setup from 4:06, then three demos - a voice-controlled browser at 5:59, AI memory at 11:33, and a YouTube predictor at 17:27. The demos take up the majority of the runtime, so expect hands-on footage rather than theory.