Aurélien Lair

@aurelienlair
5 Followers
38 Following
26 Posts

AI doesn't change anything about the fundamentals of testing. Inverting the pyramid to bet everything on E2E is going back 15 years. Essential read by Thierry de Pauw if you work in QA 📐

https://thinkinglabs.io/articles/2026/04/12/dont-let-ai-invert-the-testing-pyramid.html

#QA #SoftwareEngineering #AI

Don’t Let AI Invert The Testing Pyramid

ThinkingLabs:: Thierry de Pauw

🎙️ Next CTM Online: "Lead the testing" by Jesper Ottosen - 27 May, 7 PM Rome.

Growing into test leadership without a management title.

Join us!
https://www.continuoustestingmeetup.com/events/2026-05-27

Lead the testing

Talks:“Lead the testing” (Jesper Ottosen)

🔍 "Self-healing" in test automation sounds great on paper but does it actually hold up? Some honest thoughts on the trade-offs worth reading before you commit to it.

https://tinyurl.com/2225ccoq

#TestAutomation #QA #SoftwareTesting #QualityEngineering

My thoughts on ‘self-healing’ in test automation

From what I’ve seen over the years, tests that exercise a system through the graphical user interface are disproportionally likely to fail for reasons other than an actual, genuine product failure. Call them false positives, call them flaky tests, call them whatever you want.

On Test Automation

Stop chasing 100% coverage, start asking "what could go wrong and how badly?"

Great short read on shifting from a coverage mindset to a risk mindset in testing. A reframe every QA engineer and manager should make.

🦆 https://shorturl.at/58toQ

#QualityEngineering #Testing #RiskBasedTesting

Forget coverage; focus on risk - The Quality Duck

High test coverage can feel reassuring — until incidents happen anyway. Quality isn’t about more tests. It’s about reducing risk where it matters.

The Quality Duck - Technology Leadership founded on Quality

🤖 Continuous Testing Meetup - Apr 15 @ 7PM CEST

"LLM evaluation: A live demo" with Anupam Krishnamurthy

Can you trust an AI to judge another AI? Live demo on RAG evaluation & prompt injection attacks.

Free online event 👉 https://www.continuoustestingmeetup.com/events/2026-04-15/

LLM evaluation: A live demo

Talks:“LLM evaluation: A live demo” (Anupam Krishnamurthy)

Think you're boosting your team or holding them back? This nice article reveals subtle signs of leaders who diminish vs multiply talent. In my view it's a must-read if you want to unlock real team potential 👇👇👇

https://blog.growthinstitute.com/the-edge/5-signs-leaders-diminishers-to-multipliers

#Leadership #TeamManagement

5 Signs That Separate Multipliers From Diminishers

There's a fine line between leaders that multiply the talent of their teams and leaders that diminish it. Research from Liz Wiseman points to 5 habits that set these leaders apart.

Reminder that you can de-shittify your Google search by adding "-ai" or appending "&udm=14 to your query (see https://udm14.com/).

Duckduckgo also has noai.duckduckgo.com.

Bonus: your search comes back instantly.

&udm=14 | the disenshittification Konami code

A quick way to get an AI-free search without any extra work.

Great leaders don't always have the answers: they ask the best questions 🧐💡

As Jackie Colburn points out, curiosity fuels creativity, trust, and better decisions.

Ask more. Assume less. Lead with questions. 🤝✨

👉 https://tinyurl.com/yc6pey47

#Leadership #Curiosity #Teamwork #Innovation

Leading with curiosity: How questions, not answers, define great leaders

There’s a common misconception that leaders need to have all the answers all the time. Like in moments when a team member’s head swivels…

Medium
💡 Ash Mann highlights why digital projects fail: it’s often cultural misalignment, not tech.
⚠️ Politeness, unclear ownership & retroactive alignment = hidden risks.
🔍 Use pre-mortems, shared success criteria & alignment rituals to prevent failure.
📄 Read more: https://shorturl.at/0PUKJ
#DigitalLeadership #QA #ProjectSuccess
The illusion of alignment

Misalignment in digital projects often hides behind surface-level agreement. Cultural norms like politeness, power imbalance, and rushed decision-making lead to fragile consensus. Real alignment takes trust, open challenge, and shared definitions before delivery, not after.

Ash Mann

Just added security tests to my Juice Shop portfolio project 🚀 🚀 🚀

Now OWASP ZAP scans run automatically in ephemeral Kubernetes environments, alongside performance testing with Locust.

The full pipeline simulates a company workflow:
- merge into master ➡️ deploy to Hetzner K8s cluster
- security testing with ZAP
- performance testing with Locust
- automatic teardown of the environment after tests

Proud to use this as part of my portfolio project 🫡

Check it out 👇
🔗 https://gitlab.com/aurelienlair/juice-shop

Aurélien Lair / juice-shop · GitLab

GitLab.com

GitLab