Testing United Conference: Unleashing the Power of AI in QA
QA Engineer Iana and her colleague Sara report their insights from the latest Testing United Conference with the main topic “AI Augmented QA".
20.12.2021
written by Tatjana Statescu
Traveling for business in COVID times is chaotic at the best of times since the rules and regulations change daily. In November, circled by heightening national concern, all Austria and Germany waited with uncertainty for news of imminent lockdown and border closure. My dates were favourable, and I made use of the very last window of freedom to attend an in-person industry conference in Germany before the COVID curtain fell. Ancient, green, picturesque Potsdam, the seat of the Enlightenment was covered with drizzly rain in mid-November 2021 as I arrived. It was the host town of Agile Testing Days, a brilliant event in the Agile testing calendar that I had previously attended in 2019 and loved wholeheartedly!
When the 2021 in-person conference was announced I knew I had to be there. And thank heavens I made it there. What a refreshing change after the long, dreary 18 months of remote meetings. A big thank you to the organizers for making this happen.
I was delighted by the many side activities focused on well-being like guided city walks and morning Yoga. I enjoyed a post-prandial “fresh air & self-care” session walking around the hotel area, chatting and connecting with fellow testers. It seemed that this year’s conference contained many more talks about self-care, self-reflection, and mindfulness than in 2019. No wonder, since much of the 2 years interlude has been isolating and stressful for most of us.
There was a great variety of talks and topics. I’m writing with my impressions and thoughts rather than trying to provide a synopsis of the content.
The key point of this talk is that there is still no consensus on the definition of Artificial Intelligence because it is still difficult to define what intelligence is in humans. AI is just software so we must use our existing practices and methods. Only responsible humans can create responsible AI.
Like the AI talk, this talk approached an ethical dilemma. As software engineers, we should embed sustainability in all aspects of our work. Software is solving a lot of problems but also creates or contributes to new problems like exponentially rising energy consumption.
We can all become toxic leaders and we should watch out for the warning signs: becoming defensive, constantly comparing ourselves to others, tying our self-worth to the latest results.
The environments that create toxic leaders include a high state of stress, emotional numbness, and insecurity. There is a great blog article by Raj that tells more.
João talked about the paradox of choice: although we are inclined towards wanting more and more choice, we are usually better off with fewer choices. Limitations are good. As a tester, constraints are helpful. For example, in exploratory testing, it is good to set a time box and a focus area.
Can we use cognitive biases in our favour in our software development processes? Yes, we can.
For instance, the Zeigarnik effect, which states that people remember unfinished or interrupted tasks more readily than completed tasks, can be used to our benefit:
Ard gave us a fascinating example of a risk analysis meeting in which the walls (or virtual meeting background) are adorned with pictures of earthquakes, volcanoes, and other dangerous scenarios. The “risk thinking” of the team will thrive.
I loved the analogy of associating each learner’s type with a Pokemon character. We all are at some stage of a learning continuum. Vera offered advice and warnings for each.
Novice. Someone just starting in the field. Take care with the Novice to avoid overwhelming detail. Avoid diving into every rabbit hole.
Apprentice. Someone with a fair amount of real-life work experience. Align the choice of learning objectives appropriately with the Apprentice’s level of experience.
Crafter. Someone with broad knowledge gained from practical experience. Stretch the Crafter to step outside of their comfort zone.
Expert. Someone who knows the subject so well they define the rules themselves. Experts should be encouraged to teach, taking due care with Novices not to overload them with too much depth and detail.
Joining a long-established project with 10+ Scrum teams and 40+ microservices is hard, especially when testing processes are not well-organized and some tests overlap in different levels of the testing pyramid. Shuqi presented her success story, in which she managed to speed up testing and improve test coverage and quality.
My takeaway was that in microservice architecture when testing a single service one should try to avoid testing the business logic from other services. Contract tests are an excellent fit for that, but they come with a cost. Careful and precise definitions of requirements and communication between teams are essential.
The conference also offered multiple workshops, some of which I was fortunate to attend:
As a board game and murder mystery lover, I enjoyed this workshop. We played the board game “Black Stories”. Adapted for the Agile world, it came enhanced with ridiculous limitations and self-reflection cycles. It was fun, and I am looking forward to doing a similar workshop with our team here at AURENA.Tech.
I learned to use a WireMockProxy tool for end-to-end test automation or exploratory testing. I will certainly try this at the next opportunity.
These are only a small selection of many brilliant talks and workshops available at Agile Testing Days. To sum up, I am very grateful to AURENA for providing the opportunity to learn so many new things, to take a break from the day-to-day work in the office, and to meet people working in the same field and dealing with similar issues.
If you enjoy continuous professional development and are passionate about pursuing your career, please check out our current jobs.