Skip to Main Content
Talk Intermediate MIT License

Runtime Conditions Profiles: The Missing Contract Between Applications and Platforms

Proposal status is Review Pending
Akanksha Trehun
Akanksha Trehun
Session Description

Applications don't run in isolation. They depend on databases, caches, and APIs that affect configuration and deployment. Platform engineers fulfill them with automation, but describing the intent  what the code need is still manual, making that first deployment needlessly painful.

Runtime Conditions Profiles create the missing contract. In this session, you'll get an overview of Runtime Conditions Profiles, including how profiles are generated from code and validated against extension vocabulary. You'll see how profiles can be stored as Kubernetes CRDs or surfaced through Backstage and Headlamp, while adapters provision the external resources needed to run a codebase successfully.

Most importantly, you'll learn how to use Runtime Conditions Profiles to automate the deployment of applications to your cluster using the platform you have today no redesign or retooling needed.

Key Takeaways

Opensource review often ends with a reviewer requesting you to test the changes made in your PR locally first if not already done. This talk walks through actually doing that for a Backstage plugin that correlates Kubernetes CRDs to catalog components by git URL identity instead of fragile labels, so a single profile tracks correctly across dev, staging, and prod clusters.

The CRD itself, RuntimeConditionsProfile, isn't hand applied YAML it's delivered by a Kratix Promise, meaning the platform team defines it once and any cluster can request an instance on demand. That means Backstage doesn't care which cluster or which Promise pipeline produced the resource, only what codebase it describes.

We'll cover the work that goes through when a reviewer asks you to test it locally, no runnable app existed to test the plugin against, RBAC for the Promise's CRD was silently missing and a toolchain mismatch broke native builds across Node versions all fixed by standing up a real kind cluster, the Kratix Promise, and a Backstage instance from scratch.

Attendees leave with a pattern for identity-based CRD-to-entity correlation, how Kratix Promises fit into that picture and how to verify the work done for yourself

References

Session Categories

Engineering practice - productivity, debugging
Story of a FOSS project - from inception to growth
Talk License: MIT License

Speakers

Akanksha Trehun
Akanksha Trehun Student | St Joseph Engineering College

Akanksha is a final-year Computer Science student at St Joseph Engineering College,Mangaluru,India,and an active open-source contributor and maintainer. She's a GSoC 2026 contributor at CircuitVerse,working on LTI 1.3 integration in a large Rails codebase,and mentors and serves as a project admin for GirlScript Summer of Code,where she experiments with agent-assisted maintenance workflows.She's drawn to the intersection of automation,developer tooling, and the communities open source depends on.

Reviews

No reviews yet.