Nordic Testing Days 2025

In the era of data-driven decision-making, the performance and scalability of ETL (Extract, Transform, Load) data pipelines are crucial for managing large and complex datasets efficiently. This study explores the design and testing of an ETL data pipeline built with Apache Airflow, Python Pandas, and Pytest. Airflow orchestrates pipeline workflows, ensuring transformation dependencies and scheduling are managed correctly. Pandas handles data manipulation, offering robust tools for efficient transformations. Pytest enables a structured framework for unit testing, ensuring reliability, performance and scalability under varying data loads.

The study provides technique for data performance and scalability testing by simulating diverse workloads to identify bottlenecks and optimize execution time. Metrics such as records per second or data points per second (optionally resource utilization) are evaluated via not always provided system performance and scalability expectations.

Key Takeaways:

• Gain knowledge of creating ETL Airflow data pipeline with Python Pandas library
• Know how to test performance and scalability of ETL data pipeline by designing/executing/reporting tests via Python Pytest library
• Getting familiar with tips and tricks in Python IDE (Pycharm)

If you're working with the code, it doesn't matter if you write/test/release/review it, you need to now how securely navigate through the changes and versions. Version Control Systems are helping us to distribute our code. There is one system that leads the way for years:

Git: A tool we all use but rarely master. Why does it feel so complicated, and why is it so easy to create a tangled mess? Yet, Git remains the go-to choice for millions of engineers and organizations worldwide.

In this session, we’ll dive into Git mechanics and explore ways to make it more approachable. By shedding fresh light on common challenges, we’ll uncover solutions that make these issues a bit easier to tackle.

Ever wonder how rebase works, why merge conflicts happen, or what the heck `git bisect` is? Join me to unravel the mysteries of git and gain a deeper understanding how to use this powerful tool and not get hurt.

Key Takeaways:

• Discover how Git uses graphs and how this new context helps you understand basic commands
• Learn how to manage branches, resolve merge conflicts, and keep you changes clean
• Explore strategies to safely undo and redo the changes and how to avoid loosing your work
• Discover investigation techniques for those critical moments when “someone did something and it doesn’t work” but they need it now ;)

When I started my current assignment as a Test Manager, the clients organisation was trying to visualize how good the teams were working with test and quality by counting bugs the teams filed during development. Since then, we have thankfully come a long way.

Inspired by the Quality Criteria in the Heuristic Test Strategy Model we decided to collect and visualize test data in accordance of the Quality Criteria (QC) in ISO 25000. In this presentation I will share how we have setup all reporting, from bugs to test case results to exploratory test sessions, so that it connects to the QC's. This way, we can have an automatically updated board, created in Power BI, to visualize the current status as well as trends.

The idea is to help the organisation to understand not only what have been tested (and the results) but also what have not been tested. The aim is not to provide an objective truth but expose information so that the right questions can be asked to the development organisation. For example: Why have there not been any testing performed on usability? Do we need usability tests or is this out of scope?

Key Takeaways:

• Theory behind using Quality Characteristics and the value of tracking the affected QC for test cases, test sessions and issues, and the overview that helps the organisation to understand the current status. 
• A little bit how practical add of new fields in test management tools such as Azure DevOps Test. 
• How information can be read and visualized in business intelligence tools such as PowerBI

In the rapidly evolving landscape of software development, ensuring robust and reliable testing is critical for delivering high-quality applications. This session will explore the powerful synergy between Playwright, a cutting-edge testing framework, and AI technologies, to revolutionize UI and API testing. Playwright’s capability to handle complex scenarios across multiple browsers and platforms makes it an indispensable tool for modern development teams. By integrating AI tools such as ChatGPT, Cursor, and Claude with Playwright, we can dramatically accelerate test creation, reduce manual effort, and improve overall test coverage. This session will feature practical examples of how AI can be leveraged to: Generate efficient test cases Automate the creation of Page Objects Streamline API testing workflows A live demo will highlight the real-time integration of AI with Playwright, showcasing how to create and execute UI and API tests seamlessly. Attendees will gain tangible insights into the benefits of AI-enhanced automation, including increased efficiency, scalability, and reliability. Whether you are a QA engineer, developer, or test automation enthusiast, this session will equip you with actionable strategies and tools to optimize your testing processes. Join us to discover how AI-driven test automation with Playwright can shape the future of quality assurance.

Key Takeaways:

• AI-Powered Test Automation – Utilize AI tools like ChatGPT and Claude to automate test case generation, create robust Page Objects, and build reusable components, streamlining the test automation process.
• Integration with Playwright – Leverage AI with Playwright’s code generation (playwright codegen) to speed up automated test creation while reducing manual effort.
• Maximized Coverage & Efficiency – Enhance test coverage and productivity by combining AI-driven prompts with Playwright’s multi-browser capabilities, optimized Page Object patterns, and automated API testing workflows.
  

Many of us have to integrate third-party APIs, and we cannot trust their documentation. We also need to track when their data format changes. The same is true for integrating our own (or other teams’) microservices.

In this talk, I will present my approach to doing this safely with TypeScript and how this enables end-to-end contract testing.

This is a very practical, hands-on talk with many concrete examples that will give the audience a clear guide on how to build robust, well documented system integrations.

We leverage NPM modules to version our API definition and create a clear source of truth for all your integrations, be it a REST API, WebSocket or CoAP messages-the same principle applies.

Key Takeaways:

• Learn about TypeBox and how I use it to make integrations of third-party APIs type-safe
• See a code-example where a frontend-application runs against a mock-API using these types
• Get to know techniques to discover breaking API changes through monitoring

At Nordic Testing Days, I’m excited to share my journey at Wolt and provide actionable tips on how to unlock faster, more robust mobile releases. I’ll present four key strategies that I hope every audience member can take away and apply to their own release processes.

Over my 14-year career in IT, I’ve experienced a variety of approaches to production releases across desktop, web, and native mobile applications. For the past four years at Wolt, I’ve focused on testing and refining the release processes for our native mobile apps. When I joined, Wolt’s small consumer mobile team followed a release cycle that took 3–4 weeks to ship updates to customers. Over time, we transformed our approach, evolving our release strategy to achieve faster, more efficient, and higher-quality releases—ultimately reducing the cycle to just one week.

1. Quality Assurance to Quality Assistance
Transitioning from the traditional quality assurance model to a quality assistance approach significantly streamlined Wolt’s mobile release process. Historically, a small group of QA engineers handled release testing, but as the company rapidly scaled, this assurance-based model became unsustainable. By adopting the assistance model, we empowered teams to take ownership of their testing, involving everyone in the testing phase throughout the development cycle. This shift not only accelerated release timelines but also fostered collaboration and accountability across teams.

2. Defect Prioritization
Over the years, developers, QA engineers and stakeholders at Wolt actively addressed issues as they were discovered, whether from team feedback, support reports, or App Store reviews. While this approach showed great responsiveness, it often resulted in fixing too many issues and significantly slowing down the release process. The constant flow of new commits also increased the risk of new regressions in the application. To tackle this, we introduced a prioritization framework to rank the issues identified during the release process. This framework helps us decide which issues are critical to fix and which can be deferred, ensuring a smoother and more predictable path to production.

3. Automation
Test automation was a game-changer in shortening the release cycle. Manually running hundreds of tests was neither efficient nor scalable, so we explored and experimented with various tools and approaches to find the most effective solution.  The key was identifying the most critical parts of the app to test at both the UI and unit test levels. Additionally, we revamped our CI pipeline to better support the release process, enabling faster, more reliable builds and deployments. Automation became a cornerstone in ensuring speed and quality while reducing the manual workload on the team.

4. Be Brave and Roll!
While Android and iOS ecosystems have their own default release processes, we discovered that following them too cautiously can be both time-consuming and unnecessary. Most issues are typically caught early in the rollout, allowing for faster adjustments if needed. Emphasizing the importance of feedback loops, tools like Crashlytics and other third-party providers play a crucial role in monitoring releases and ensuring a smoother journey. Sometimes, taking a confident approach and accelerating the rollout can save time while maintaining quality.


In this talk, I’ll share how we transformed Wolt’s mobile release process—from 3–4 weeks to just one week—by adopting a Quality Assistance model, prioritizing defects effectively, leveraging automation, and embracing faster rollouts with robust feedback loops.

Key Takeaways:

• Testing is not just QA’s responsibility, it’s a team effort.
• Defect prioritization: making the right fix at the right time.
• Release automation is more than just unit and UI tests.
• Releasing native apps isn’t always about following Google and Apple’s standard flow.
  

In an era where speed and seamless interactions define digital success, website performance can make or break the user experience. Core Web Vitals (CWV), a standardized set of performance metrics defined by Google, have become essential for measuring web quality from a user-centric perspective. This session will introduce the three CWV—Largest Contentful Paint (LCP), Interaction to Next Paint (INP), and Cumulative Layout Shift (CLS)—highlighting their direct impact on both user satisfaction and business outcomes.

Designed specifically by a software tester with extensive experience in CWV, this talk will uncover why CWV are critical in  web application quality strategies, demonstrating how they expose performance bottlenecks often overlooked during development and testing. Attendees will gain useful insights into measuring and analyzing these metrics using tools like Lighthouse, PageSpeed Insights, and Real User Monitoring (a technique that captures actual user interactions to assess real-world performance).

Beyond the theory, this session will provide actionable guidance on integrating CWV into your testing pipelines, reporting performance issues with more detail and clarity, and collaborating effectively with developers to drive meaningful improvements. Whether you're new to performance testing or seeking to expand your expertise, you'll leave equipped with practical techniques to embed performance accountability into your quality culture.

Key Takeaways:

• Core Web Vitals Help to Quantify the Web Performance Quality
  • Metrics like LCP, INP, and CLS directly measure the user’s experience with loading speed, responsiveness, and visual stability.
  • Good performance enhances user satisfaction, retention, and even SEO rankings.
• Software Testers Play a Critical Role in Helping to Discover Performance Issues
  • Testers can identify performance risks early by integrating Core Web Vitals checks into their test plans, CI/CD pipelines, and monitoring.
  • Performance testing isn’t just a developer’s task—collaboration between testers and developers drives a shared quality mindset.
• Practical Tools Make Performance Testing Accessible to Testers
  • Tools like Lighthouse, PageSpeed Insights, and Real User Monitoring simplify measuring Core Web Vitals.
  • Testers can track, report, and prioritize performance issues alongside functional defects for a more complete quality strategy.

Modern test automation frameworks often struggle to keep pace with the rapid evolution of dynamic web applications. Frequent UI updates lead to brittle tests and broken locators when UI has strict design of dynamic web element due to Regulatory or Compliance Restrictions, creates bottleneck in agile workflows and CI/CD pipelines. Healenium, an advanced AI-driven self-healing tool, addresses this challenge by seamlessly integrating with Selenium to dynamically repair broken locators during run.

This presentation delves deep into Healenium’s architecture, which combines machine learning models and contextual analysis to detect, predict, and heal test failures caused by UI changes. We will explore its ability to maintain a centralized locator repository, track locator evolution, and apply predictive healing strategies to ensure test continuity. Real-world case studies will showcase how organizations using Healenium have achieved over 50% reduction in test maintenance efforts, minimized downtime in CI/CD pipelines, and enhanced test reliability.

Key topics include:

• The principles of self-healing test automation and how Healenium leverages locator similarity algorithms.
• Step-by-step integration of Healenium into existing Selenium frameworks with minimal disruption.
• Comparative analysis: traditional test maintenance vs. Healenium-enabled self-healing.
• Best practices for maximizing the potential of Healenium in complex, large-scale test environments.

Healenium is redefining the future of test automation by empowering teams to deliver faster, smarter, and more reliable software.

This session is designed for advanced test automation engineers, QA architects professionals seeking cutting-edge solutions to address test instability in dynamic applications by reducing the test maintenance overhead using AI and machine learning. Attendees will learn how Healenium improves test resilience, optimizes CI/CD pipelines, and enables QA teams to focus on writing new tests rather than repairing broken ones.

Key Takeaways:

• Discover the Power of Self-Healing Automation
• Reduce Test Maintenance Overhead
• Future-Proof Your Automation Suite