Тёмный

V0 + Claude: A GAME CHANGER for Building AI Apps 

Cole Medin
Подписаться 4,9 тыс.
Просмотров 15 тыс.
50% 1

Artificial Intelligence is no doubt the future of not just software development but the whole world. And I'm on a mission to master it - focusing first on mastering AI Agents.
In this video I show how to use a SUPER powerful AI combo - V0 + Claude - to build a Next.JS frontend for an AI app in just minutes. The app I build a frontend for is the RAG AI Agent I built earlier on my channel with n8n and Supabase that many of you got excited about, so this is a good extension of that walkthrough!
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Everything I go over in the video, including the n8n workflow, all the prompts, and the Next.JS code with my ShadCN components can be found here:
github.com/col...
Here is everything you need for this walkthrough:
- V0: v0.dev/chat
- Claude 3.5 Sonnet: claude.ai
- NPM installed on your computer: www.npmjs.com/
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Join me as I push the limits of what is possible with AI. I'll be uploading videos twice a week - Sundays and Wednesdays at 7:00 PM CDT!

Опубликовано:

 

18 сен 2024

Поделиться:

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист
Посмотреть позже
Комментарии : 35   
@frustasistumbleguys4900
@frustasistumbleguys4900 10 дней назад
V0 + Cursor with Claude 3.5 is INSANE.
@ColeMedin
@ColeMedin 9 дней назад
I agree! I have been checking out Cursor on top of V0 and Claude and I'm definitely going to use it to build on this example in the future
@serendipity-108
@serendipity-108 9 дней назад
What sort of integrations have you found useful? Or are you just using v0 first for design/UI components?
@ColeMedin
@ColeMedin 8 дней назад
@@serendipity-108 Yes v0 is really just used for frontend components! But I do use Claude for a lot of other things, both to help me code my AI Agents and as the LLM my AI Agents often use. Is that get at what you are asking? Otherwise feel free to clarify and I'd love to get more specific!
@Arkfiit
@Arkfiit 6 дней назад
You got my subscription and more views Just keep teaching me I'm always grateful Ready to make millions ❤
@ColeMedin
@ColeMedin 6 дней назад
Thank you man, that means a lot!!
@sahul879
@sahul879 3 часа назад
I rarely subscribe to any channels, but you've definitely earned mine.
@entranodigital
@entranodigital 8 дней назад
Amazing Content, Love the depth of this. This combo is seriously fire
@ColeMedin
@ColeMedin 8 дней назад
Thank you - that means a lot to me!! I'm glad you also think this combo is incredible :)
@Keon12211
@Keon12211 8 дней назад
Amazing video! Congrats on the rising views on your channel. I am a total beginner (never coded before) so I dont know about which softwares/apps to use. Could you tell me what you have used in this video? Like visual studio code? Java Script? Node? Also would it be possible to see how you installed your 2 triggers? Thanks!
@ColeMedin
@ColeMedin 8 дней назад
Thank you very much and welcome to the world of coding! So for V0 and Claude those are both just in the browser. Then, once I created a project on my computer in the second half of the video, I used: - VS Code for my code editor - Node + NPM to create the Next.JS project and bring in my V0 component - Typescript as the programming language (Typescript is an extension of JavaScript) Then in my previous video, titled "This RAG AI Agent with n8n + Supabase is the Real Deal", I covered this n8n workflow in more detail so you can see there how I set up the triggers! The only one not included there is the webhook trigger, but the chat trigger and the Google Drive triggers I cover in more detail. Let me know if you have any more questions!
@Keon12211
@Keon12211 7 дней назад
@@ColeMedin Thanks! It did, I got alot further. Can i ask how you setup the "edit fields" between the chat message received and Webhook? I cant get it to connect both to the RAG AI Agent. I appreciate the effort!
@Beloved_Digital
@Beloved_Digital 8 дней назад
Thank you sir. Please can you make a roadmap video or begginer video that will be a perfect guide for a beginner who wants to build AI powered apps like this. Thank you
@ColeMedin
@ColeMedin 8 дней назад
Of course! And that's a great idea! I am actually planning a longer form video where I have a roadmap of beginner AI projects to more advanced AI projects (with in-depth guides for each). Hopefully that is the kind of thing you are looking for!
@Beloved_Digital
@Beloved_Digital 8 дней назад
@@ColeMedin yes sir. Enough information to guide me build useful AI apps and agents . Thank you so much!
@morena-jackson
@morena-jackson 8 дней назад
@@ColeMedin This would be awesome!
@frankm7307
@frankm7307 10 дней назад
Yes, coding with Claude is fast Yes, you should totally make like a landingpage or just the first setup of a webapp Yes, it outputs code with seriously little errors and can resolve them quickly afterwards, when returning the errors. One very big thing to note, though, is that if your project becomes somewhat large, Claude often gets lost. Suddenly unable to remember how to read documents that yoi share with it, eventhough it claims to have read it. Constantly forgetting the details of the code it wrote, as quickly as the next prompt. If you really want to make amazing things on AI, you need to figure out how to work around these hurdles. If you do, you'll have a lot of fun. If you get complacent, you will have a bad time.
@jarad4621
@jarad4621 10 дней назад
Yeah some kind of modular framework to work with it I don't think you need to give all your code each time to work on specific areas of a large app I'd you are super clear on the instructions like explain the context of the larger app in prompt without giving it the code as well maybe
@ColeMedin
@ColeMedin 9 дней назад
Yeah that's a very good point and something I've experienced myself when working with larger codebases! It's for sure a limitation of LLMs that I foresee getting better overtime with models that can handle larger contexts better. But as @jarad461 said, you typically don't need to give your entire codebase to solve specific problems, so generally you'll piecemeal what you give the LLM to help you solve a specific problem or create a specific component. I do look forward to the day though where LLMs can easily ingest entire codebases and not get confused
@drumbassclassics953
@drumbassclassics953 9 дней назад
The reason it forgets is its memory capacity. That has and will continue to grow. It’s better than OpenAI.
@ishaanpotnis
@ishaanpotnis 9 дней назад
Can you make a tutorial using v0 chat and Firebase Cloud Functions and Realtime Database for backend please?
@ColeMedin
@ColeMedin 9 дней назад
That's a great idea! I actually have used Firebase a lot in the past so I could definitely see myself making this video in the future! Thanks for the suggestion :)
@chadpogs7973
@chadpogs7973 10 дней назад
Wow!!
@jarad4621
@jarad4621 10 дней назад
Awesome
@ColeMedin
@ColeMedin 9 дней назад
Thank you!
@blarvinius
@blarvinius 9 дней назад
Any tips: I already HAVE a React app.... Is there a decent workflow to redo the (hideous looking but working) UI with v0?
@ColeMedin
@ColeMedin 9 дней назад
Great question! All the components generated in V0 you can copy and bring directly into your existing React app - you don't actually have to do the npx install command for the ShadCN component (that's just for convenience). So basically you can paste in your existing component(s) you want to improve for the V0 prompt, and then copy what it creates and either replace the code in your existing file or make a new component and use that instead. Hopefully that helps!
@blarvinius
@blarvinius 9 дней назад
Cool! 🐟
@marcc0183
@marcc0183 9 дней назад
bro how i can upload/update the rag from the web app in nextjs?
@ColeMedin
@ColeMedin 9 дней назад
With this setup, the knowledgebase is updated directly through adding/updating files in a Google Drive folder. But I am certainly planning on making a video in the future where I build a frontend to manage the knowledgebase with file uploads!
@kunalr_ai
@kunalr_ai 9 дней назад
how to deploy it
@ColeMedin
@ColeMedin 8 дней назад
Great question! There are a lot of ways to deploy a Next.JS app - the easiest being using Vercel. nextjs.org/learn-pages-router/basics/deploying-nextjs-app/deploy If you want to be super cost efficient when heavily scaling a Next.JS app Vercel can get kinda pricy, but in that case you could look into self-hosting your Next.JS app on a cloud instance with a service like DigitalOcean.
@fokinsansan
@fokinsansan 8 дней назад
this shitcis awesome
@ColeMedin
@ColeMedin 8 дней назад
Much appreciated! 🔥
Далее
The Story of Next.js
12:13
Просмотров 577 тыс.
Why OpenAI o1 has Changed the Game for AI Agents
12:45
Просмотров 2,1 тыс.
Is Computer Science still worth it?
20:08
Просмотров 117 тыс.
Why did OpenAI move from Next.js to Remix?
10:41
Просмотров 123 тыс.
Build anything with v0 (3D games, interactive apps)
9:05
This RAG AI Agent with n8n + Supabase is the Real Deal
16:27
Full Stack Developers will take over. This is why.
11:26