Quality Engineering meetup #12

PLEASE NOTE THE ADDRESS CHANGE! We're now at w3hub.berlin

Qase is presenting the twelfth Meetup on Quality Engineering in Berlin.
As usual, we’re bringing you three deep-dive talks, but also something extremely new and exciting — food testing activity. Don't worry, we won't be the test subjects, we will feed the spectrometer and see if our tests pass or fail :)

AGENDA

1. Jaideep Naik, Senior Software Engineer: The Migration Hell: From Ghost Inspector to Cypress

Jaideep will share his lessons from a 12-month test automation migration from Ghost Inspector (low-code) to Cypress (code-first). Limited planning led to flaky tests, slow CI pipelines, and lost trust in automation. Covers the key challenges — skills gap moving to JavaScript, tightly coupled tests, inefficient Page Object Model, UI-based auth bottlenecks, static test data, no coding standards — and the solutions that fixed them: phased migration, page-based POM with standardized locators, API-based auth, dynamic test data, and PR reviews from the start.

2. Radik Zagirov, Founder @ Agentiqa: Quality Is Binary: Why Your Coverage Metrics Don't Matter After the First Escalation

Radik argues that shipping code has become essentially free — AI writes code, CI/CD deploys it. But quality hasn't kept up. The bottleneck has shifted to people and processes, and business perceives quality as binary: everything works, or QA failed. One escalation erases months of coverage improvements and process building. Radik will share his understanding, that the answer isn't better testing but rather better communication. How to explain QA's value in the language of money, risk, and speed, and how to build "opportunity tunnels" — systematic channels that make QA's impact visible before an incident forces the conversation.

3. Anupam Krishnamurthy, consultant: Ensemble Testing of an LLM Application

Earlier in November, Anupam demonstrated how we could test LLMs by using the LLM-as-a-judge approach. In that session, somebody asked, how can we trust an LLM to judge another LLM?

In this session, we put LLM judges to the test. We'll do some live testing to have an LLM judge evaluate an existing RAG application.

We'll then put up the results on the big screen and take a look together, to verify if our LLM judge is working as expected. In other words, everybody in the room will judge the judge.

Can an LLM judge survive the scrutiny of a hundred pair of expert eyes? Come and find out in this live ensemble testing session.

4. Sebastian Kasanmascheff, CEO/Owner @ Altum Solutions:
Wetware QA Jumpstart: Hands-On Spectrometer Demo

Did last meetup’s “Food QA: Testing Tasting” session whet your appetite for wetware testing? Well here is an invitation to turn theory into practice with a live showcase of the device that taste-tests like a human, namely, the spectrometer.

Take a leap from the familiar realms of software and hardware into the exotic world of wetware: translate between computer bits and chemical bonds, complement your programmer hoodie with a lab coat, and let us test what is in our food:

- How much protein is really in your protein powder?
- Can you trust the fat content values on your milk packaging?
- What ingredients are hiding in your favourite fast food meal?

Learn how to answer such questions by joining the Altum Solutions team for an end-to-end spectroscopy workflow: from sample preparation to ML/AI-enhanced scientific analysis. This same wetware QA pipeline empowers sustainable lab automation across Big Food, AgriTech, and Pharma 4.0.

BYOB — Bring Your Own Bytes/Bites