Program Wednesday March 3, 2027

09.30 Registration
 

10.00 Workshops round 1

Lidia Barkanova Vladislav Kutsevalov
A1 - Who Tests the AI? Upgrade Your Test Automation with AI Evals

expect(answer).toContain(“refund”) – PASSED ✅
Actual result: “Sorry, we never give a refund.”

Test automation is great at checking exact results. But more and more products integrate generative AI, and it never answers the same way twice. Your assertion passes when it should fail, and you can’t read thousands of answers by hand. Our answer: AI evals, automated checks that score AI answers. We built them for our own product. In this hands-on workshop, you will learn to add them to Playwright tests and leave ready to test AI at scale.

Lidia Barkanova

Senior Software Engineer, Infobip

Vladislav Kutsevalov

Staff Quality Engineer, Infobip

Workshop for beginners

Prerequisites – Your own laptop with VS Code (or another IDE), internet access, and permission to install software. We will send setup instructions in advance.

 


Paula Bassagañas Odena
B1 - One Agent To Watch Them All: Building an AI Test Monitor

Are you tired of keeping one eye on your CI/CD pipelines? Checking builds, digging into logs, and chasing failing commits? Talking to AI won’t help much; we need actions, not words.

In this hands-on tutorial, you’ll build an AI CI/CD monitor using Model Context Protocols (MCPs). MCPs act like a modern USB port for AI, turning existing APIs into safe, testable contracts the agent can use. Your agent will: watch CI/CD in real time; read build outcomes, failing commits, and test metadata; analyse failures to surface root causes and impacted areas; and send structured email/notification reports.

Paula Bassagañas Odena

Functional Architect, Boehringer Ingelheim / Testing Fantasy

Workshop for experience attendees

Prerequisites – A Laptop to access the browser-based environment.
Basic knowledge of API concepts and Python is helpful, but not strictly required. We will provide comprehensive step-by-step guides and code snippets.


Sara Martinez
C1 - Testing for Trust: Essential Security Testing Across the SSDLC

In the AI Era, security is no longer optional. In this hands-on workshop we will learn how to integrate security testing across the entire software lifecycle. Participants will explore threat modeling, secure design, and practical testing techniques (including SAST, DAST, CI/CD automation, and risk reporting) so quality and security evolve together. Through real examples and exercises, attendees will learn how to embed security into agile delivery and strengthen the resilience of the software they build.

Sara Martinez

QA Lead, S2GRUPO

Workshop for beginners

Prerequisites – Bring your laptop and have Python and Docker installed.


Stefan Schenk
D1 - MCP in Action: Turning APIs into AI-Powered Test Data Generation

In a microservice environment the number of services and therefore api’s can become quite large in number. To make these api’s easily accessible and available as agent tool we can setup an MCP server. By creating the right tools we give ourselves the ability to use the services to our advantage without needing to know exactly what api needs to be called and what data is needed.
Wouldn’t it be nice if you could just type “I need a travel plan for a party of four traveling by bike in Limburg for five days, with hotel bookings for each day.” and your complete test data is created for you?

Stefan Schenk

Senior Test Automation Specialist, deTesters

Workshop for experience attendees

Prerequisites – Bring your own laptop, this will be hands-on An IDE, eg. IntelliJ. Basic knowledge of Typescript or Python (tbd), Claude Code, OpenCode or equivalent.
You need some programmer experience, because we will be writing code. And knowledge about REST api’s is preferred. 


Jan Molak
E1 - Beyond Writing Test Scripts: Refactoring to Value

AI-assisted development is making test code cheaper to produce. But more tests don’t automatically create more value. In this hands-on workshop, you’ll learn how to build reusable interaction models, compose them across testing levels, and turn test execution into useful evidence. Using a realistic test suite, you’ll refactor duplicated automation step by step and explore how this architecture changes what happens when AI can generate tests at scale.

Jan Molak

Principal Software Engineer, T. Rowe Price

Workshop for experience attendees

Prerequisites – Participants should have practical experience with automated testing and be comfortable reading and modifying test code. No prior knowledge of Serenity/JS, the Screenplay Pattern or AI-assisted development is required. Participants should bring a laptop with a modern web browser and be able to install, run, and modify a provided Node.js project. The workshop repository and setup instructions will be provided in advance.


13.00 Lunchbreak and visit exhibition
 

14.00 Workshops round 2

Ester Greenwood
A2 - Thinking Inside of the Box: White-Box Testing at PickleCorp

Congratulations, you’ve been hired! PickleCorp welcomes you to the PickleFactory development team. We make the grandest gherkins, the perkiest peppers and other snackable vegetables that are delicious, nutritious and not at all suspicious. Well.. at least when we get our machines working properly. We have 100% green checkmarks on our tests and yet we are finding too many errors. It is your task to set up better unit testing and to analyse the technical integration. The health and safety department also requires reporting. Test the journey of our data points from machine to readable report.

Ester Greenwood

Test Engineer, Sopra Steria

Workshop for beginners

Prerequisites
– A laptop, though the analysis part of the exercises will also be in the reader
– Participants will be sent a list of instructions to install VS code and other tools like node.js and playwright (I don’t have a definitive list on this yet)
– Participants will have to clone the codebase from GitHub. They can check if it works by running the program.
Some kind of test automation experience in an IDE is preferable but not mandatory.


Tim Lolkema
B2 - Agentic Testing & AI Skills

In this session, you’ll develop a feature using agentic engineering and have it autonomously tested with playwright-cli. You’ll evaluate test results and evidence to determine when you can trust the agent, build your own Agent Skill for writing E2E tests, and refine and integrate it into an existing development workflow. Finally, you’ll explore when Agentic Testing adds value and when scripted automation remains the better choice.

Tim Lolkema

Quality Lead, deTesters

Workshop for beginners

Prerequisites
https://timlolkema.nl/preparation.html
– Have a laptop with installation rights
– Install Node.js
– Install the Playwright CLI
– Clone the demo webshop
– Have a coding agent that supports skills
The workshop is beginner-friendly, so no prior experience with AI agents or Agent Skills is required. However, some basic technical knowledge, such as installing software and working with development tools and git, is helpful.


Marco Maes Erik Haartmans
C2 - Testing Native and Webview Mobile Apps the Playwright Way

If you’ve automated mobile apps with Appium, you know the pattern: selectors that break with every app update, and sleep calls scattered through your code because waiting for the right condition never quite works. Mobilewright applies the same locator, auto-wait, and Page Object Model concepts that made Playwright popular to real Android and iOS apps, but it’s young and doesn’t cover everything Appium does. In this hands-on workshop you set up a project, build a small Page Object Model, write a webview round trip test, and hear honestly where the approach still falls short.

Marco Maes

Test Automation Architect, Polteq Test Services B.V.

Erik Haartmans

Test Automation Architect, Polteq Test Services B.V.

Workshop for experience attendees

Prerequisites – A laptop with Node.js 24+, Android Studio with an emulator set up, and a code editor. I’ll send a short setup guide beforehand so we don’t burn workshop time on installs. Basic TypeScript or JavaScript. Prior Appium or Playwright experience helps but isn’t required, I’ll explain the concepts as we go.


Craig Risi
D2 - Design, Attack, Defend: Engineering Guardrails for AI

What happens when an AI agent has access to your source code, development tools, sensitive data and deployment pipelines—and then makes the wrong decision? In this hands-on workshop, participants will take on the challenge of designing an AI system that is capable enough to be useful but constrained enough to be trusted.

Craig Risi

VP Software Engineering, Entersekt

Workshop for experience attendees

Prerequisites – Participants will need a laptop, a GitHub account, and the ability to install packages directly on their machine.
Some basic knowledge of existing qulaity guardrails and AI is helpful – as we won’t be explaining this.


Anaïs van Asselt
E2 - When AI outruns QA: Build the Review Harness

How do you keep quality high when AI is accelerating development? You build a review harness: a safety net of deterministic checks and AI judgment. In this hands-on workshop, you start by doing a reviewer’s job yourself, reading real pull requests and deciding what matters, then build that judgment into an AI review skill and close the feedback loop with an evaluation set. You leave understanding when to write a deterministic check and when a change needs AI judgment instead, ready to apply that to your own codebase.

Anaïs van Asselt

Senior QA Engineer, Choco

Workshop for experience attendees

Prerequisites
– Bring a laptop with Git and Node.js installed.
– Download the prepared repository before the workshop.
– Participants do not need their own access key for the AI service.
– Live runs use a temporary project key behind a small workshop server. The key is never shared and is deleted after the workshop.
– Prepared evaluation results are available if the connection fails. Participants can continue the exercise, but cannot run their own new instructions offline.
Participants should be able to read a small frontend or backend diff and understand the purpose of unit, integration and end-to-end tests. They do not need to be application developers or AI experts.


17.00 Drinks at the exhibition