Last night I attended the BrowserStack Meetup Group — Amsterdam, an evening dedicated to one of the most consequential shifts in our field: how artificial intelligence is reshaping quality assurance. The meetup ran from 6:00 PM to 8:00 PM CEST at Capital C Diamantbeurs, in the heart of Amsterdam.
The world of QA is changing fast. AI promises huge efficiency gains in test automation, but it also introduces new failure modes — model hallucinations, agents drowning in excessive context, brittle tests that look green but verify nothing. The agenda was built around exactly that tension: how to get the leverage of AI without the pitfalls.
The Agenda
The evening was structured around two complementary talks, with plenty of room for networking:
- 6:00 PM — Welcome and Networking
- 6:10 PM — Just Enough Context: Teaching Claude to Test with Playwright — Eleonora Belova
- 6:50 PM — Break
- 7:00 PM — Generative AI in Quality Engineering: Beyond the Hype — Robert Jadoenandansing
- 7:40 PM — Dinner and Networking
Talk 1: Just Enough Context — Teaching Claude to Test with Playwright
Eleonora Belova (QA Automation Engineer at Virtual Vaults) tackled one of the subtlest problems in AI-assisted testing: context. Give Claude too little and the tests are shallow; give it too much and it hallucinates or loses the thread. Her talk focused on precisely controlling the context you feed an LLM so it produces reliable, hallucination-free Playwright tests.
This is squarely in the territory of context engineering — the discipline of deciding what an agent can see, and what it should ignore. For anyone building AI-driven test automation, the lesson is that the model is only as good as the window you give it.
Talk 2: Generative AI in Quality Engineering — Beyond the Hype
Robert Jadoenandansing (Director at TestBotics Academy) took a wider lens, looking at the real-world impact of Generative AI across the entire QA lifecycle — from test design through execution to debugging. The “beyond the hype” framing mattered: the goal was actionable knowledge about where GenAI genuinely helps and where the limitations, risks, and ethical considerations bite, especially in enterprise QA environments.
Why This Matters
Both talks converged on a theme I keep returning to: AI does not replace the QA engineer’s judgment, it changes the shape of the work. The new skills are about scoping models, validating their output, and knowing when a generated test is trustworthy. That is as much a prompt-engineering and platform-engineering problem as a pure testing one.
Below is a photo walkthrough of the evening.
Welcome & Networking
Welcome & Networking at Capital C Diamantbeurs.
The Talks in Pictures
The two talks: Eleonora Belova on Claude + Playwright, and Robert Jadoenandansing on Generative AI in QA.
Dinner & Hallway Track
Dinner and hallway-track conversations on where AI-driven quality engineering is heading.
