I’m a third-year Digital Systems and Design student at Aalto University. I’m interested in domain modeling, constraint systems, and building production software that handles real-world complexity.
Designed and built a production course discovery & planning platform used alongside SISU at Aalto.
- Domain-driven core (Plans, Blocks, Schedule Pairs, Decision Slots)
- Deterministic constraint-combination & ranking engine (no fake "optimal" schedules)
- Dual-dataset architecture (active + append-only historical), disjoint by construction
- Snapshot “patch layer” for missing instances — non-canonical data only graduates via deterministic CI backfill
- TypeScript (Express) backend, Svelte frontend, PostgreSQL (RLS) with DB-enforced invariants
- Offline-first PWA: course browsing works without internet via versioned cached datasets
- Fully deployed with environment isolation + health-verified releases
🔗 Live: https://sisukas.eu
🔗 Code & docs: https://github.com/kctong529/sisukas
If you care about systems that make trade-offs explicit, survive real-world constraints, and are actually deployed and maintained, feel free to take a look :)

