sinodo A membership platform built on the idea that AI doesn't have to be so complicated

sinodo is where you find someone who already used AI in a business like yours. No endless courses. No smoke. Just what works.

in development 2026
Built with Next.js 15 React 19 TypeScript Tailwind v4 Clerk Supabase Cloudflare Vercel Claude Code

There's too much noise about AI
and very little real experience being shared

The problem isn't that information about artificial intelligence is scarce. It's that there's too much of it. 40-hour courses, Twitter threads promising results in 7 days, influencers selling the same prompt reformatted every week.

What doesn't exist — or barely exists — are people with real businesses explaining exactly how they used a specific tool, what worked, what failed, and why. No personal brand filter. Not optimized for virality.

sinodo started from that observation: useful knowledge about AI applied to real businesses isn't in the courses. It's in conversations that haven't happened yet — between people who already did it and people who are about to.

The bet:
tiered access, real community, no fluff

sinodo works with three access levels. The free tier gives access to public content and the base community. The member tier unlocks resources, tools and deeper conversations. The founder tier is for those who want direct access, no intermediaries.

Free Public content + base community
Member Resources, tools and full forum
Founder Direct access, no intermediaries

No Stripe yet. The decision was to build the functional platform first and validate that the model makes sense before activating payments. That order matters.

The challenges that weren't
in the documentation

Building a membership platform seems straightforward until you start connecting the pieces. Each integration has its own layer of complexity, and some technical decisions that seem minor end up being the ones that consume the most time.

01
Next.js 15 + Clerk

Breaking changes in middleware behavior between versions, with no clear documentation for auth-at-the-edge projects.

02
Tailwind v4

Breaks with the previous config approach. The community hadn't caught up with the official docs yet.

03
Clerk + Supabase RLS

Connecting two different identity systems has more steps than it looks. Every misaligned detail returns the same generic error.

04
React Server Components + Client Components

Correctly splitting what runs on the server and what runs on the client, especially in a dashboard with complex state.

The stack and why
each piece is where it is

There was no predetermined stack. Each technology came in because it solved a specific problem better than the alternatives at that moment.

Framework
Next.js 15 (App Router) React 19 TypeScript
Styles
Tailwind v4 PostCSS Custom CSS tokens
Auth
Clerk JWT + RLS
Database
Supabase PostgreSQL Row Level Security
Infrastructure
Vercel Cloudflare (DNS/Edge)
Development
Claude Code

The fonts (Corben for headings, Inter for body) come from Google Fonts and reinforce the visual identity: something between editorial and functional, without falling into the generic. The green (#2A5C45) and cream (#F5F0E8) define the two extremes of the color system.

What already exists
and what's left to connect

sinodo isn't a prototype. It's a functional platform with built pages, working authentication and a database with real structure.

Built
  • Landing with hero and tiers section
  • Blog with listing and individual posts
  • Dashboard with Server Component + Client UI
  • Community and contact pages
  • Clerk authentication (sign-in, sign-up)
  • Supabase tables with RLS: profiles, posts, resources, forum, tools
  • Proposal generator as the first dashboard tool
In progress
  • Deploy on Vercel
  • Full dashboard tools connection
  • Testing with real authenticated users
  • Full mobile responsive
  • Activate payments (Stripe)

Why this project
matters beyond the code

sinodo is technically the most complex project I've built. It involves more layers, more integrations and more product decisions than any tool I've launched before.

But what I want to demonstrate isn't technical complexity. It's that today it's possible to build a complete membership platform — with authentication, database, dashboard and community — without an engineering team. With the right tools and the willingness to understand what you're building.

If you have a community you want to turn into a platform, or a membership idea you haven't launched because it seems too hard to build, it probably isn't anymore.