Senior Cloud & DevOps Engineer
I build cloud and data-platform automation, Python services, and developer tools. My background runs from hands-on, 24/7 telecom datacenters to Databricks on GCP, with a focus on making production work easier to deploy, troubleshoot, and recover.
At work, that means Terraform, Python, FastAPI, Databricks Apps, Lakeflow Jobs, and CI/CD. In my public projects, I work on practical problems around data processing, recovery, and coding-agent workflows.
| Project | What I built |
|---|---|
| Subzero | Subtitle cleanup, validation and translation with a persistent SQLite worker queue, restart recovery and regression tests for cleanup failures and file permissions. |
| Agent Safepoint | Rust workspace checkpoints with content verification before restoration. Restores files without rewriting Git history. |
| CRM ETL Toolkit | CSV/XLSX contact cleaning and validation with a local FastAPI workflow. A synthetic pipeline example runs without a CRM account. |
| Backchannel | A Python standard-library bridge for reading local coding-agent sessions and queuing replies, with token authentication, upload limits and workspace-scoped discovery. |
- Databricks CLI #7036 (open): prevent repeated schema/catalog updates caused by properties managed outside a bundle.
- qBittorrent #25048 (open): validate packaged macOS applications, including dependencies, application startup and artifact checksums.
- Railpack #772 (open): preserve the pinned Gradle version instead of silently resolving a different release.