22 Verwendung insgesamt•00 Gesamte Likes•00 Anteil insgesamt•00 Gesamt eingespart•
vor 6 Monaten
en
Männlich
Jung
Bildung
Erzählung
Mittlere Stimmlage
Ruhig
Entspannt
Bedächtig
Klar
Neutraler Ton
Professionell
Geschichtenerzählen
Geschmeidig
Bug fixes and improvements
During this sprint, we saw a significant increase in views of the KAT documentation, as well as deeper engagement with the framework from different teams.
Because KAT is being used more actively, we focused a lot on stability, bug fixes, and framework improvements. Our main goal was to make the framework more reliable, easier to use, and easier to maintain.
Improved KAT codebase
One of the important improvements was the cleanup of the KAT codebase.
We removed hundreds of lines of duplicated Bash scripts and replaced them with reusable Python components. This helps us reduce technical debt and makes the framework much easier to test, maintain, and extend in the future.
As a result, the code is now cleaner, more structured, and better prepared for future improvements.
Simplified Kubernetes configuration and better support for new environments
We also improved the Kubernetes configuration.
As you may know, the new version of KAT does not use the old kat-config anymore. Instead, it uses the new kat-config, which we have already deployed to all engineering and production environments, except for a few special environments that will be added later.
This means that teams and customers using the new KAT will be able to run it whenever they need, without depending on the old configuration approach.
In addition to that, we simplified the Kubernetes setup. Now it is much easier to add and configure new clusters. We also added support for small-bang deployments.
These changes should make the experience smoother and more flexible for teams working with new environments.
Java base image
We also expanded the KAT framework by introducing a new Docker base image for Java applications.
The idea is very similar to what we already have for KAT Newman. Teams will be able to take the base image, add their own tests, and run them in the KAT framework.
This will make it much easier to integrate Java-based applications and will help us support a wider range of technologies in our ecosystem.
We are almost done with testing this feature, and we expect to make it available soon.
Switch to AI-driven development
As part of the support for the Wolf project, we also started moving toward AI-driven development.
We began using AI tools in our daily development process to work faster and more efficiently. One of the tools we started using is OpenSpec, which helps us describe, manage, and implement changes in a more structured way.
This is still a new approach for us, but we believe it will help the team improve productivity and make development more consistent.
Rollout of KAT 2.0 to pre-production
Finally, we started implementing the new KAT 2.0 for several services, including FDE, DC, and Valuation.
Together with these teams, we are preparing to move to pre-production in the next sp.
This is an important step toward wider adoption of KAT 2.0 and will help us validate the framework in more real use cases before expanding it further.
Closing
To summarize, this sprint was mostly focused on improving the quality, stability, and usability of KAT.
We reduced technical debt, simplified configuration, added support for Java applications, started using AI-driven development practices, and prepared KAT 2.0 for rollout to pre-production.
These changes should make KAT easier to maintain for us and easier to use for other teams.
