Manish Mukherjee

Did I need all this tech for a marketing site?

No. But here's why I did it anyway.

Development Timeline

Day 0
2 hrs

AI-assisted scaffolding

v0.dev + Claude Code

Day 1
8 hrs

Engineering refinement

Testing, polish, production

Traditional
40-60 hrs

Without AI tooling

Industry standard estimate

Total: 10 hours vs 40-60 hours — 75-85% time reduction

Stack Breakdown

Testing Suite

This is the part AI can't do well (yet). And it's what separates a demo from production.

15
Unit Tests
2
E2E Tests
100%
Type Safety
CI/CD
Pipeline
AA
WCAG
# Test output
PASS
Test Suites: 2 passed, 2 total
Tests: 15 passed, 15 total
Snapshots: 0 total
Time: 2.1s

Production Checklist

Security
HTTPS enforced
Input validation (Zod)
Supabase RLS
Rate limiting ready
Email auth (SPF/DKIM/DMARC)
Performance
Server Components
Image optimization
Code splitting
Edge caching
Monitoring
Vercel Analytics
Error tracking ready
Uptime monitoring
Log aggregation

Reality Check: AI vs Engineering

What AI Gave Me
90% of UI components in first pass
Responsive layouts out of the box
Consistent styling patterns
Boilerplate elimination
Occasional hallucinated APIs
Inconsistent error handling
What Engineering Added
Comprehensive test coverage
Proper error boundaries
Accessibility audit & fixes
Security hardening
Performance optimization
Production monitoring setup

Want this approach for your project?

I combine AI-accelerated development with 20+ years of engineering discipline to ship faster without sacrificing quality.

Complete Intake Form