From side hustle to shipping: ArchitectGBT now has New UI & Smart Features! 🚀
Hey everyone! 👋
A few weeks ago, I shared ArchitectGBT and got amazing feedback. You asked for improvements, and I shipped them.
Here's what's new:
What is ArchitectGBT (quick recap):
An AI model recommendation engine that stops you from wasting hours comparing Claude vs Gemini vs GPT. Describe your use case, get the perfect model in 60 seconds with exact costs and production-ready code templates.
🎨 What I Just Shipped:
Landing Page Redesign:
Complete mobile-responsive overhaul (tested on all devices)
Sticky navigation with smooth scroll to sections
Cleaner, more focused design inspired by successful SaaS products
Added "How It Works" section (community feedback!)
Enhanced Features Section:
Expanded from 3 to 6 detailed feature cards
Real-Time Cost Analysis - see exact pricing per 1M tokens
Production-Ready Code templates (26+ and growing)
Smart Model Matching based on your requirements
Live Model Database with 15+ latest models
Time-saving calculator (hours → seconds)
AI Models Showcase:
Now prominently displays all 15+ models we support
Organized by provider: OpenAI (GPT-4.5 Turbo, GPT-4o, etc.), Anthropic (Claude Opus 4.5, Sonnet 4.5, etc.), Google (Gemini 2.5 Pro, 2.0 Flash, etc.)
Shows what each provider is best for
Weekly database updates
Technical Improvements:
Fixed CVE-2025-55182 security vulnerability
Centered pricing card titles (small details matter!)
Optimized images and performance
Better mobile breakpoints (sm/md/lg)
📊 The Numbers So Far:
15+ AI models in database (OpenAI, Anthropic, Google)
26+ production-ready code templates (TypeScript, Python, cURL)
3-step process: Describe → Analyze → Deploy
60-second average recommendation time
Featured on Product Hunt! 🎉
💡 Why I Built This (The Real Story):
I wasted 6+ hours last month researching which AI model to use for a side project. Cost per token? Context window? Rate limits? Which provider? Should I pay for GPT-4 or try Claude?
It was overwhelming. The documentation is scattered across 3+ provider sites, pricing calculators are confusing, and nobody tells you which model is actually BEST for your use case.
I built ArchitectGBT because this should take 60 seconds, not 6 hours.
🔥 Looking for Beta Testers:
I'm bootstrapped and building lean, so your feedback directly shapes what ships next. Here's what I'd love to know:
Does the new UI make sense? Is anything confusing?
Would you actually use this for your next AI project?
What's missing? What features would make this a must-have?
Is the pricing fair? (Free tier gives you 3 queries to try it out)
Try it: ArchitectGBT: Find Your Perfect AI Model in 60 Seconds | Product Hunt
Solo Founder Journey:
Building this while working a 9-5 and managing family time. Some nights I code until 2am, other nights I get zero lines shipped. It's messy, it's hard, but shipping these updates and getting your feedback makes it worth it.
Thanks for all the support on the last post. You pushed me to ship faster and better. 🙏
I'm here all day to answer questions, take feedback, and chat about AI models, solo founding, or whatever!
Pravin
P.S. - If you've struggled with choosing AI models for your projects, I'd love to hear your war stories in the comments. Helps me understand what to build next! 🚀
Replies