Role overview
Develop the low-level systems that make a database store, index, and recover data reliably. The work centers on the database engine itself, including storage, caching, transactions, concurrency, and performance at large data volumes.
Responsibilities
- Design, profile, and implement core features such as storage formats, index structures, and recovery paths.
- Investigate concurrency defects, race conditions, and lock contention.
- Analyze CPU use, memory allocation, off-heap memory, and disk I/O to improve latency.
- Review designs and code with attention to correctness, data integrity, and maintainability.
- Help shape architectural choices as the database engine evolves.
- Write thorough tests for complex kernel behavior.
Requirements
- Systems programming experience and comfort working close to the hardware; Java is used extensively, while experience in C++, Rust, or Go can also transfer.
- Understanding of concurrency concepts such as threads, locks, and atomic operations.
- Knowledge of data structures, cache locality, and algorithmic complexity.
- A pragmatic approach to engineering, including testing edge cases.
- Collaborative communication and willingness to support technical decisions with evidence or benchmarks.
Nice to have
- Experience with database internals, storage engines, or file systems.
- Knowledge of JVM internals, including garbage collection, JIT compilation, or off-heap memory.
- Experience with profiling tools such as async-profiler, JFR, or perf.
- Familiarity with large, mature codebases.
Benefits and work setup
- The team is distributed between two European cities and discusses database research to assess ideas for practical use.
- Regular time is set aside for individual experiments, prototypes, or tooling improvements.