“Building large software remains hard. And it always will, because our ambition will forever outstrip the metal.”
Distributed systems → concurrency, queues, storage, networking
Cloud Native → Kubernetes, containers, Linux, observability
AI / LLM systems → agents, structured output, evaluation, safety
Developer infrastructure → CI/CD, automation, testing, tooling
Application engineering → APIs, full-stack systems, real-time software
I also spend a lot of time working upstream rather than only building isolated projects.
| Ecosystem | What I've worked on |
|---|---|
| openSUSE / Uyuni | reliability fixes, SSL/certificate diagnostics, FQDN verification, UX fixes, documentation, dependency/security updates |
| openSUSE / Open Build Service | validation hardening, malformed-input handling, crash fixes, race-condition handling, and performance improvements |
| Rage | WebSocket stream lifecycle APIs and OpenAPI-related work in a modern Ruby backend framework |
| Talawa | GraphQL/DataLoader performance work, cache invalidation, test coverage, developer tooling, and rootless container development |
| CNCF ecosystem | contributions across Koordinator, Volcano, Chaos Mesh, and KubeEdge |
| Observability tooling | Superlog tests, CI, telemetry utilities, and reliability-focused changes |
That work has taught me an important lesson: a good patch is not just code that works locally; it is code that fits the architecture, survives edge cases, and is easy for the next engineer to trust.



