✨ AI Powered
SaaS for Designer Growth
How do you objectively assess designer growth? I built a platform to answer that question.
BloopScore·SaaS Platform·Designer assessment tool
The Problem
- Designers have no objective way to measure their skills.
- Portfolios are subjective. Interviews are biased.
- "Am I junior or senior?" — a question with no clear answer.
- Traditional SaaS dev? That's 6+ months with a team.
What I Did
Founder
- Architected entire system: 152 files, 33 API endpoints
- Integrated Claude API for dynamic lesson generation
- Built real-time matchmaking with race condition handling
- Wrote 52 planning documents for methodical AI-assisted development
Key Wins
- Born from a real leadership need: assessing designer growth objectively
- AI-powered skill assessment and personalized learning paths
- 18K LOC built in 22 days with AI-assisted development
How It Works
flowchart TB
subgraph User["👤 User Journey"]
A["Sign Up\nGoogle OAuth"] --> B["Assessment\n5 Skill Categories"]
B --> C["Results\nSeniority Level"]
C --> D["Learning Path\nAI Generated"]
end
subgraph AI["🤖 AI Engine"]
E["Claude API"]
F["Lesson Generator"]
G["Weekly Auto-Gen\nGitHub Actions"]
end
subgraph Gamification["🎮 Engagement"]
H["XP System"]
I["Medals & Badges"]
J["Leaderboard"]
K["Real-time Matches"]
end
subgraph Data["💾 Data Layer"]
L[("PostgreSQL\n+ RLS")]
M["Redis\nRate Limiting"]
end
C --> E
E --> F
F --> D
G --> F
B --> H
D --> I
K --> J
H --> L
K --> M
Quick summary complete
Want the full story?Keep reading
📖 Full Case Study
🎯 The Challenge
Managing complexity at scale: 152 TypeScript files, 13 DB migrations, real-time matchmaking with race condition handling. 52 planning documents ensured methodical workflow: plan → document → implement → test.
✅ The Solution
AI-first architecture with Claude API for dynamic lesson generation. Microeducation system adapts content based on user performance. GitHub Actions for weekly auto-generation. Modern stack: Next.js 15 + React 19 + TypeScript strict + Supabase + Vercel.
BloopScore dashboard and assessment interface
The Numbers
18,000 Lines of TypeScript
152 files, 50+ React components, 33 API endpoints, 13 database migrations. All built in 22 effective working days.
90+ Unit Tests
Comprehensive test coverage for auth system and core functionality. Production-grade quality from day one.
Key metrics from the build
Technical Stack
Frontend
Next.js 15, React 19, TypeScript strict mode. Glass morphism design system, avatar system, gamification with XP and medals. Server Components + Server Actions for optimal performance.
Backend & Infra
Supabase (PostgreSQL + Auth + RLS + Realtime), Google OAuth, Resend emails, Upstash Redis for rate limiting, Vercel CI/CD. 8 external service integrations, all type-safe.
Modern production infrastructure
Impact
Production-Ready in 22 Days
From zero to 18,000 LOC production SaaS in 22 working days. 33 API endpoints, 90+ tests, 8 service integrations. Ready for real users.
85% Time Reduction
Traditional team estimate: 6+ months. Actual delivery: 22 days. AI-assisted development without sacrificing code quality or test coverage.
Project Learnings
AI Multiplies, Doesn't Replace
AI as a multiplier enables building complete SaaS products at speed. Architecture decisions, data modeling, and UX strategy still require human judgment.
Documentation is Process
52 planning documents aren't an afterthought—they're how you build complex systems methodically with AI. Plan → document → implement → test.
Interested in working together?
Looking for leadership roles at product companies building design infrastructure at scale.