From Waiter to AI Architect: 10,000 Hours of Operational DNA
Most developers learn to code, then try to understand business. I did it backward—and that's my unfair advantage.
The Floor
I started as a waiter at Muze Cafe. Not because I wanted a career in F&B, but because I needed to survive. I was a kid from a village without electricity until Class 5. No computer until college. No formal education in anything related to technology.
But something happened on that floor that would change everything.
I watched the chaos. Orders getting mixed up. Inventory counts being off. Payroll taking 40+ hours every month because everything was manual. Staff training inconsistent because it relied on whoever had time to explain things.
"I didn't see a restaurant. I saw a system full of inefficiencies waiting to be optimized."
The Climb
Over 10 years, I moved from waiter to operations manager. Managing 50+ staff. Overseeing $1M+ P&L. Leading digital transformation initiatives. But here's what most people don't understand:
I wasn't just managing. I was documenting every pain point.
- Which POS system features were actually used vs. which were ignored?
- Where did inventory discrepancies always occur?
- What questions did new staff ask repeatedly during training?
- Which manual processes took the most time with the least value?
I became obsessed with understanding not just what was broken, but why it was broken and how it could be fixed.
The Code
Then I discovered programming. Completely self-taught. No bootcamp. No CS degree. Just relentless curiosity and YouTube tutorials at 2 AM after closing shifts.
The difference was: I already knew exactly what problems needed solving.
Most developers learn React, then wonder what to build. I had a backlog of 10+ years of problems.
The Systems
My first real project was PUNCH-CLOCK: a biometric attendance and payroll automation system. Why? Because I'd spent hundreds of hours manually processing payroll and knew every edge case, every error pattern, every compliance requirement.
The result? Reduced payroll processing from 40 hours/month to 2 hours/month. 95% time reduction. But more importantly: it actually worked in the chaos of a real restaurant environment.
Then Vine_AI: a wine pairing recommendation system. Because I'd watched premium wine bars struggle to train staff on complex flavor profiles and pairing rules. 300% ROI increase.
Then FLAIR AiAI: multilingual conversational AI for staff training. Because I'd seen the same training questions asked hundreds of times in multiple languages. 40% faster onboarding.
The Advantage
Here's what 10,000 hours of operational experience gives you that a CS degree doesn't:
- Reality-tested requirements: I don't build features users "might" want. I build solutions to problems I've personally experienced.
- Edge case awareness: I know what breaks systems in production because I've seen it break in real environments.
- Business context: I understand P&L implications, not just technical architecture.
- User empathy: I've been the person using clunky software during a dinner rush. I design accordingly.
What This Means for You
If you're hiring me for an F&B tech project, you're not getting a developer who needs to "learn the domain." You're getting someone who lived it for a decade.
If you're hiring me for any automation project, you're getting someone who understands that "efficiency" isn't about elegant code—it's about giving people their time back.
Bottom line: I don't just build systems. I build systems that work in the messy reality of actual operations.