Cloud, platform and software engineering for systems that need to be reliable, observable, secure and operable.
This GitHub organization hosts public engineering work maintained by the team behind Omnivya.
Our open-source work focuses on practical infrastructure and platform problems: understanding dependencies, operating Kubernetes and cloud environments, improving CI/CD security, making infrastructure costs explicit, and building tooling that removes operational friction.
- Cloud & infrastructure — Azure, hybrid architectures, infrastructure automation and cost visibility.
- Kubernetes & platform engineering — operators, workload lifecycle, GitOps and platform automation.
- SRE & observability — Prometheus, OpenTelemetry, dependency signals, diagnostics and reliability engineering.
- Security & software supply chain — SBOM, secret detection, CI/CD security and policy enforcement.
- Developer & operations tooling — focused tools designed to be understandable, testable and useful in real environments.
Discovers external dependencies used by Kubernetes workloads, builds blast-radius context and exports provider-neutral SRE signals across cloud and SaaS systems.
Queries Azure retail pricing to make infrastructure cost assumptions explicit, comparable and reproducible.
Runs self-hosted Azure DevOps agents on Kubernetes and adjusts capacity according to pipeline demand.
Generates SBOMs, scans repositories for exposed secrets and evaluates software supply-chain policies with Dependency-Track.
Provides Compose-style orchestration for Apple's container runtime on macOS, including dependency-aware startup, health checks and operational commands.
Installs Smallstep CLI reproducibly in GitHub Actions with CI, CodeQL and release automation.
We prefer explicit architecture, observable behaviour and documented limitations over opaque automation.
Public repositories should make it possible to understand:
- why the project exists;
- what problem it solves;
- how the main components interact;
- what security and operational assumptions it makes;
- what is stable, experimental or intentionally unsupported.