12 nodes / 10+ apps
Operate platforms
Bare-metal Kubernetes, storage, observability, networking, identity, secrets, certificates, scanning, and GitOps delivery.
- Kubernetes
- FluxCD
- Ceph
- Grafana
- Prometheus
- Loki
Stack & Engineering
My working toolkit.
Sushil Poudel’s platform, data, and delivery toolkit—Kubernetes, FluxCD, Ceph, Python, PostgreSQL, Amazon Redshift, and GitHub Actions.
What the stack proves
The page is organized around repeatable engineering capabilities first, then the tools that support them.
200 GB+ weekly
Validate data
Healthcare data transformation with SQL analysis, client control checks, and delivery-ready output validation.
- SQL
- PostgreSQL
- Amazon Redshift
- File Mapper engine
- Scrub engine
1,700+ pages
Automate delivery
CI/CD, research tooling, OCR processing, and XML validation with clear boundaries between build, parse, and data-quality checks.
- GitHub Actions
- Python
- Quarto
- XML
- CI/CD
Complete working toolkit
My toolkit reflects work across an on-premise platform, healthcare data engineering, and university research. The categories below keep the broader skill set available without making the first screen a long technology inventory.
Platform & infrastructure
Linux, Ansible, Kubernetes, FluxCD, Helm, and Ceph. At jaraco, my work includes a 12-node bare-metal Kubernetes cluster, storage integration, and GitOps delivery. I use infrastructure tooling to connect the runtime, its services, and the release process.
Observability, networking & security
Grafana, Grafana Alloy, Prometheus, and Loki for visibility; Flannel and Cilium for cluster networking; Keycloak, SonarQube, HashiCorp Vault, cert-manager, Let’s Encrypt, and Trivy for identity, code quality, secrets, certificates, and scanning. These are the integrations covered by my platform work.
Data engineering
PostgreSQL, SQL, Amazon Redshift, Apache Kafka, Confluent, File Mapper engine, and Scrub engine. My healthcare data work combined transformation with validation against client controls. My wider interests include event-driven data systems and the operational questions around streaming.
Programming & research tooling
Python, Bash, HTML, CSS, JavaScript, Quarto, and XML. At the University of Münster, I used Python and Quarto for research web tooling and processed OCR output into conformant XML documents.
Delivery & collaboration
Git, GitHub, GitHub Actions, and FluxCD. My work spans automated application delivery, frontend CI/CD, agile sprints, and translating client requirements into implementable tasks.
Current direction
I’m developing an AI-integrated data mesh system at jaraco GmbH. The work is ongoing; my focus is on bringing platform engineering and data operations closer together.