Ronald Veth

I'm a test automation architect who believes great frameworks aren't born from proof-of-concepts. They're designed from architecture. After 12+ years of building, breaking, and rebuilding test solutions across finance, energy, healthcare, and logistics, I've seen what happens when a PoC quietly becomes the production framework: blind spots that nobody understands, changes that break everything, and teams that lose trust in their own tests.

My job is to prevent that. At Cerios, I advise organisations on test automation strategy, design frameworks that scale, build reusable open-source tooling, coach teams, and deliver hands-on. I'm equally comfortable in a boardroom explaining why architecture matters as I am writing Playwright fixtures in TypeScript.

I hold an HBO Informatica degree from Hogeschool Rotterdam and serve on the program committee of Test Automation Days, where I review abstracts and help shape the conference agenda.

Every framework I build starts with a question: what does the architecture demand? Not "what's the quickest test I can write." Understanding the system under test, including its APIs, data flows, authentication, and integration points, determines the framework's shape. The building blocks come from that analysis, not from a generic template.

This philosophy drives everything I do: from the SPARTA framework with its standardized building blocks, to the open-source packages I publish on npm, to the way I coach teams to think about test design.

I started in C#, moved through Java, and landed firmly on TypeScript as the language for modern test automation. Here's why:

• JSON-native: APIs speak JSON. TypeScript understands it natively. No serialization libraries, no mapping layers.
• Zod for contracts: Generate schemas from OpenAPI specs, validate API responses at runtime with full type safety. When the API changes, the tests break at exactly the right place.
• Utility types: Pick, Partial, and Omit let you write type-safe assertions on exactly the fields you care about, ignoring the rest without losing confidence.
• Playwright: Best-in-class browser automation, built for TypeScript. Fixtures for dependency injection, custom engines for component testing, trace viewer for debugging. It's the full package.

TypeScript gives test automators the expressiveness of a scripting language with the safety net of a typed one. That's a rare combination.

I'm a fan of AI in test automation, but with a clear boundary. AI is a copilot, not the pilot. The moment you can't explain what your test asserts and why, you've lost control.

Here's the reality: AI generates compilable code fast. But compilable isn't the same as correct, maintainable, or meaningful. Without structure, AI produces spaghetti that passes today and mystifies everyone tomorrow.

That's exactly why I built SPARTAFICIAL, an AI layer that works within the guardrails of a proper framework. The building blocks are made by test automators who understand code. AI uses those blocks to generate tests against your specific domain. Structure first, speed second.

AI also changes how we think about tooling. Every package I build is designed with token awareness in mind: small, focused packages that accelerate test creation. When AI generates a test, it should be able to use these building blocks instead of regenerating everything from scratch each time.