CatalystOS

Catalyst OS is a full-stack life optimization platform built on Next.js 15, React 19, and Supabase β€” deployed on Vercel with auto-deploy from the main branch.

The platform serves three layers of free and premium tools across five life dimensions (Body, Mind, Heart, Wealth, Spirit):

106 interactive calculators that run entirely client-side with real-time calculation, save-to-account functionality, and PDF export. Each calculator uses established formulas (Mifflin-St Jeor for BMR, Navy method for body fat, Epley for 1RM) and renders with a consistent CalculatorShell pattern providing breadcrumbs, dimension theming, and structured FAQ JSON-LD.

225 gamified learning modules organized in a skill tree with 6 difficulty tiers per dimension. Each module delivers a three-unit structure: research-backed deep dive articles rendered with a custom markdown-like component supporting callout blocks, concept flashcard decks with flip animation and mastery tracking, and scenario-based guided practice rounds. Progress persists via localStorage for guests and syncs to Supabase for authenticated users. The XP and rank system (Blind β†’ Decoded) provides tangible progression across 1,500+ flashcards and 600+ practice scenarios.

A premium AI journaling engine (The Mirror) powered by Claude Sonnet 4 with RAG context gathering β€” pulling from journal entries, calculator snapshots, profile data points, and extracted insights to deliver personalized, evidence-based coaching across 5 conversation modes.

206+ AI-generated articles with 13 structured SEO metadata fields each, auto-generated OG images via a dynamic /api/og endpoint, and IndexNow integration for near-instant search engine discovery.

A complete consulting services section with 7 industry-specific landing pages, ProfessionalService + LocalBusiness JSON-LD schema, OfferCatalog structured data, and FAQPage markup.

An e-commerce layer combining physical merchandise via Printify fulfillment with digital product delivery through tokenized download links, all processed through Stripe with a unified checkout API.

Technical highlights that leverage Vercel’s platform:

Server Components as the default rendering strategy β€” no client JavaScript ships for static content pages. Client Components are isolated to interactive elements: calculators, flashcard decks, practice rounds, GSAP animations, and the chat interface.

ISR for blog content with 10-minute revalidation. Route-based code splitting keeps the 225-module Decode system performant despite its data size (content is split across 30+ data files per dimension to manage bundle size).

Dynamic OG image generation via /api/og using @vercel/og for social sharing thumbnails with dimension-specific theming.

Edge-compatible API routes for calculator saves, newsletter subscriptions, and content likes.

The entire platform is built and maintained by one person β€” a chemical engineer turned AI systems architect. The codebase serves as a living portfolio for an AI consulting practice, demonstrating production-grade Next.js architecture, Supabase integration patterns, and automated content pipelines.

Stack: Next.js 15, React 19, TypeScript, Tailwind CSS 4, GSAP 3, Supabase (PostgreSQL + pgvector), Stripe, Resend, Claude API, Vercel.

1 Like