Here's dilemma. claude.ai project (thanks for Brandon's tutorial!) doing half-decent job in creating functional backend code. But every time I want to change or improve styles - it generates bunch of changes in HTML + JS files and as a result - code blows up and it takes me hours to fix it for new styles to work. When claude.ai introduces bug in JS + HTML combo - it is having really hard time finding the cause and fixing it (and so as ChatGPT btw) and thus my ideas of any visual improvement are becoming big waste of time for no improvement in backend functions. I also tried Brandon' suggestion of providing it a snapshots of desired UI look - and i did it with the same result - big changes in HTML+JS+CSS -> code blows up -> after 2 hours of trying to find an issue - I reverted to previous working version with less nice UI. Luckily I setup github project with frequent updates. My current project is Web app that allows to CRUD entries in Vector DB and then observe how these entries helps LLM's to deliver answers. It is LangChain-based RAG customized for certain specific use cases. Backend functions are all working. On a frontend it involved few pages, with nav menus, tables, inputs, options, forms, view/update buttons etc. - which ideally i want to look cool, elegant and professional. So I think that maybe the right approach should be slightly different: - I should pick a solid UI framework for Web app myself at the very beginning of the project - and guide claude.ai to use it. - Then any adjustments in a future look and feel will be minor (say framework already supports dark- and light modes, presets, different table styling, etc...) - Then i continue focusing on backend functions moving forward. How does it sounds? My problem is I am totally clueless in UI/frontend frameworks and thus inviting a discussion on this subject - with hopefully interesting outcomes that all will benefit from!

Here is a great video of Greg Isenberg and Ras Mic demoing the new bolt.new. https://www.youtube.com/watch?v=lDMhK8DamuE

Hey everyone! 👋 Today, I’m super excited to share a sneak peek of my latest video, where I walk you through the ultimate AI tech stack for building MVPs (Minimum Viable Products) lightning fast. If you’ve been trying to get your AI projects off the ground or want to quickly showcase proof-of-concepts to your clients, you’re in for a treat. 🚀 Why This Stack? In the upcoming video, I break down a powerful combination of Claude, V0 and Cursor. These tools, when used together, help you build beautiful and functional Next.js MVPs in record time. The Stack Breakdown: - Claude – Acts as the brains of the operation, helping you brainstorm ideas, outline, and strategize. Imagine Claude as your project’s foundation, setting the direction before you even write a line of code. - V0 – This tool is your MVP powerhouse. It can crank out Next.js components like a pro and streamline server-side logic. Perfect for getting the main chunk of your app up and running fast. - Cursor – Finally, when it comes to the fine-tuning, Cursor jumps in to polish the details, handle data cleanup, and make those last-minute tweaks that turn a good MVP into a great one. ✨ Real-World Use Case: Real Estate Offer Letter Generator Let me give you a sneak peek of a real-world example. In the video, I walk through how I built a Real Estate Offer Letter Generator in just a couple of hours. Here’s what the app does: 1. Allows real estate agents to input client details and upload documents like client conversations and property specs. 2. Automatically generates offer letters based on the uploaded information. 3. Provides agents with an interface to tweak values like the offer price, export the final document, and send it off to clients with just a few clicks. The coolest part? What usually takes hours of manual work was done in minutes using this AI tech stack! 😲 👷‍♂️ Building the Pyramid: From Concept to MVP in No Time When building out MVPs, I like to think of it as constructing a pyramid:

Hey! I just got a great question from Tom about what AI subscriptions I’m using, and since I just wrapped up recording that section of my course, I figured I’d share it with you all. https://www.loom.com/share/ea9ac81827a8445292e8d4704d30be11?sid=690b31f4-fc48-4906-9532-752428fc8fad Here's what I cover in the video: - What AI tools and subscriptions I’m currently using as a developer. - How I integrate GitHub Copilot, ChatGPT, and Claude in my workflow. - The reasoning behind my setup, including API integrations with Cursor. I’d love to hear what subscriptions you’re using and why! If you’ve got any questions or want to share your setup, let me know.

A great YouTube video on why the vlogger Dave Ebbelaar moved from VS Code to the new Cursor IDE.