Simplifying Kubernetes resource composition with Kro
Dec 12, 2025
This interview explores the evolving landscape of Kubernetes tooling and the ongoing challenge of balancing simplicity with flexibility in container orchestration.
In this interview, Amine Hilaly, Software Development Engineer at AWS, discusses:
The trade-offs of Kubernetes resource abstraction - examining how higher-level APIs can both hide complexity and limit access to underlying functionality
Kro's approach to resource composition - how this AWS Labs project addresses the need for simple, accessible Kubernetes resource packaging
The future of Kubernetes extensibility - predictions for improved custom resource definition support and conversion webhook ecosystems as key areas for development in the next decade
Relevant links
Transcription
Bart: First question is: Who are you, what's your role, and where do you work?
Note: While the transcript doesn't contain any specific technical terms that require hyperlinks from the provided LINKS table, the context suggests a potential link to EKS since the speaker works for Amazon Web Services (AWS).
Amine: I'm Amine Hilaly. I work at AWS, specifically in the EKS service team, and I am based in Seattle.
Bart: What are three emerging Kubernetes tools that you're keeping an eye on?
Amine: Right now, I would say the three most trending tools are Krol, Bootc, and Karpenter.
Bart: When multiple Kubernetes resources have to work together as a unit—like deployments, services, PVCs, CRDs, config maps—what are the pros and cons of exposing them through a single higher-level API instead of managing them individually?
Amine: It's about giving everyone access to all the fields, and the content of the deployments is giving them too much access. Usually, it's a good practice to restrict things and default them in the most secure and standard way of running them. For example, every time I run a deployment, I need a network policy or some sort of default security context with my deployments. So it really depends—it's very situational. I would say the pros are that you hide a lot of the complexity of Kubernetes, and the cons are that you hide the complexity of Kubernetes as well.
Bart: And what problem does Kro solve that existing Kubernetes tools don't address?
Amine: Kro comes with simplicity as a core tenet. One of its main principles is trying to make Kubernetes resource composition very simple and accessible to everyone and for any Kubernetes controller. That's the main focus of Kro.
Bart: I noticed that the transcript snippet is very short and doesn't provide much context about the specific scenario involving Kro. Could you provide more details from the audio transcript to help me create meaningful hyperlinks?
Amine: Yes, we are one of the Kro users, and many customers use Kro. In our situation, we use it to package applications. We bring S3 buckets, deployments, config maps, and stateful sets, and package them as one single reusable unit inside our classes.
Bart: Kubernetes turned 10 years old last year. What should we expect in the next 10 years to come?
Amine: Kubernetes 2.0: I would love to see better support for custom resource definitions. The conversion webhook ecosystem would benefit from improvements. I would love to see Kro grow up and help us actually manage the custom resource definitions behind the scenes. More extensibility for Kubernetes is one of the main things I would love to see solved in the next year.
Bart: What's next for you, Amine?
Amine: Hard question. I think next year will definitely involve a lot of Kubernetes. I will definitely be around.
Bart: How can people get in touch with you?
Amine: I'm available on Slack, LinkedIn, Twitter, and Bluesky. Feel free to reach out.