I built this as an open-source platform for running, tracking, and comparing quantum experiments through a web UI, API, CLI, and SDK.
It includes async job execution, experiment/result persistence, provider abstractions, a demo dashboard, and a local development setup with Docker or lightweight local mode.
The goal was to build something more operational than a one-off script or isolated demo: a control-plane style developer platform for quantum experimentation.
It’s still an early-stage project, but the core execution pipeline is working end-to-end. I’d really appreciate feedback on the architecture, developer experience, and whether the problem framing makes sense.
Main focus: system integrity, not feature expansion.
Key improvements:
- enforced idempotent job execution (row-level locking, no double runs) - fixed queue processing guarantees (ack correctness + recovery) - ensured end-to-end consistency across API, worker, and storage - aligned all clients (SDK, CLI, UI) with backend capabilities - removed misleading or incomplete APIs
Also cleaned up developer experience: → full local environment runs with a single command → no manual setup, no hidden steps
Validated the system using quantum algorithms:
- Bell / GHZ (entanglement) - Grover (probability amplification) - Deutsch–Jozsa (function classification)
The interesting part wasn’t building features — it was making sure everything behaves correctly under real execution conditions.
That’s where most complexity lives. https://github.com/mareksuchodolski12-hash/kwantowy
#distributedsystems #backendengineering #quantumcomputing