One engineer. End to end.
I'm Rafał Pakuła. Round Panda Technologies is a one-person company registered in Poland — and that's the design, not a stage. Every system on this site was scoped, built, deployed, and answered for by the person writing this sentence.
Ten years of delivery, back-end-first: PHP and Drupal since 2016, Go and Kafka since 2024, TypeScript and React across every decoupled front end in between. I work remotely from Poland (CET) with Nordic and European clients, in English and Polish.
What I optimize for is boring in the best way: systems that keep running after I leave the room — documented, observable, tested, and upgraded on a schedule instead of in a panic.

Ten years, in order.
- 2015drupal.org member
- 2016PHP + Drupal delivery begins
- 2018PIM technical lead · building materials
- 2022headless brand platforms · Drupal + Next.js
- 2024Go + Kafka in production
- 2026today — still one engineer, all in production
Ways to engage
Project delivery
A fully integrated estate or a standalone app — scoped and built end to end, solo. From architecture through deployment to handover documentation. Fixed scope or phased; you always know who's accountable.
Long-term stewardship
A maintenance-and-support retainer for the platform you already run: upgrades on schedule, incidents answered, integrations kept alive. My longest engagement of this kind is in its eighth year and counting.
Specialist reinforcement
Embedded in your team for the hard part — the migration, the integration, the decoupled seam. The engagement is always direct, with you — even when an agency is already running the project.
The obvious question: what happens if you depend on one person?
You should ask it. Three honest answers.
Built to survive me. Everything I ship is documented, tested, observable, and on standard stacks — any competent team can take it over. Handover documentation is a deliverable, not a favor.
The track record is continuity. My longest engagement is in its eighth year and still running in production; the oldest systems under my care date back to 2016.
Capacity is managed, not improvised. I don't overbook. Absences are planned and announced ahead. And when a build is bigger than one person, long-trusted collaborators join under Round Panda — with one point of accountability, still mine.
Describe the system, not the job title.
Two or three sentences about what won't flow, won't upgrade, or doesn't exist yet — and where it runs. I'll reply with an honest read: what I'd do, what I wouldn't touch, and whether I'm the right person for it.
USED ONLY TO REPLY. STORED ON ROUND PANDA INFRASTRUCTURE IN THE EU.