Why I'm doing this

Each year I write down the plans I find interesting but don't build. Some because of cost, others because they feel too big, others because they'd be 'someday' projects. Over the past ten years that list grew. A coaching platform, a productivity tool, an insights dashboard, and at the top: a fully automated trading system.

All those ideas used to require a team, a budget, and months of work. Today they require something different: a sharp vision, a good prompt, and an AI that writes the code.

I find that fundamentally exciting.

What changed recently

I've built several platforms recently. Multi-tenant SaaS, development pipelines, agents talking to each other, full cloud deployments. Not as hobby projects but as running systems for real clients.

Most was written by AI. I was involved less and less. For this trading project I'm taking one step further: not a single line of code from my own hands.

My role is architect, reviewer, decision maker. AI does the executing work. What surprises me most: how few people realize this. Even software developers sometimes make the wrong choices, miss the best tools. The learning curve to do this well isn't steep, but it is specific. You have to know what you don't know.

Why a trading system now

Investing has been my passion since my student days. Buffett, Lynch, Graham, Greenblatt, Terry Smith, the classic investors who wrote about disciplined systems. I've spent years thinking about how to automate those lessons. A model that scans the stock universe every morning, scores on proven factors, decides when to enter, when to exit, and executes without emotion.

The technical implementation of such a system used to require a quant team. Today it's a challenge that's achievable in eight weeks for one person with the right tools.

The experiment

In eight weeks a system will run that daily scores 575 stocks from the S&P 500, AEX and AMX, makes decisions on buy, sell, hold or trim, executes orders via Interactive Brokers, and documents everything publicly on playitsmart.nl.

Starting capital: €10,000 of my own money. Maximum acceptable annual loss: 15%. Live from late June 2026.

My commitment to myself and to anyone watching: I don't write code. All development happens with AI. I design, review, test, decide. AI does the rest.

What I'll share here

Weekly updates on LinkedIn. In depth on this site: the technical architecture, the factor selection, the mistakes, and once it's live the performance. Fully transparent.

My goal isn't just returns. Returns would be a nice bonus. My goal is to learn and to show how far AI-driven development has come.

Who this is for

You're in the right place if you're wondering how far you can get today without programming yourself. Or if you're curious about systematic investing and want to see how a multi-factor model behaves in the real world.

The best is if you find both questions interesting. Then you've come to the right audience.

Follow on LinkedIn for weekly updates. Or subscribe to the RSS feed if you want everything in one place.

More to come.