Database kernel developer at CUBRID · Spatial data nerd · Building things that store, search, and reason over data.
I'm a database kernel developer at CUBRID, an open-source RDBMS, where I spend most of my time deep inside the storage and query engine. My academic roots are in spatio-temporal databases and 3D GIS, and I enjoy following how classic DBMS internals keep evolving as new workloads show up.
I like problems that sit at the intersection of:
- 🗄️ DBMS internals — storage, transactions, concurrency, query processing, indexing
- 🧭 Geo-spatial intelligence — 3D GIS, indoor space modeling, spatio-temporal data
- ⚙️ Systems performance — locking, parallelism, SIMD, profiling
- 🌱 Open-source software — community-driven engineering, reproducible benchmarks
- 🏢 CUBRID — Database kernel developer · contributor since CUBRID 11
- 🎓 Pusan National University — Integrated M.S./Ph.D. program in Spatio-Temporal Databases & GIS (2015–2018, coursework completed, degree not conferred)
- ✍️ Notes & write-ups on hgryoo.dev
Database internals Storage & buffer management · transactions, MVCC, locking · query processing · indexing (B-tree, multi-dimensional, ANN) · recovery & logging
Concurrency & systems performance Multi-threaded engine design · lock contention analysis · vectorization · CPU- and cache-level performance analysis · workload-driven micro-benchmarking
Spatial & spatio-temporal data 3D geometry processing · spatial indexing · indoor space modeling · OGC / IndoorGML standards
Open-source DBMS engineering Build systems & dependency hygiene · regression test infrastructure · reproducible benchmarks · release & community workflow
Software engineering culture & process Code review practices · branching & release strategies · developer experience inside long-lived C++ codebases · how teams sustain quality at scale
Mostly C++/C in the engine, Java/Python around the edges, plenty of SQL.
- CUBRID — storage engine, lock manager, CAS server, indexing
"Databases are just a very opinionated way of agreeing on the truth."