YDB is a distributed, fault-tolerant, open-source relational database that combines high availability, scalability, and performance. Together with its accompanying stack of serverless technologies, YDB enables the creation of a data storage and processing system for the most complex and demanding applications.

YDB is also a platform for building infrastructure components. We offer users not only reliable transactions on top of tables but also ready-made solutions based on the engine itself: persistent queues, federated queries, network disks for virtual machines (Yandex Network Block Store), and others.

What tasks await you

Optimization of key YDB components Due to the specific design of the distributed system, we use an actor system and network libraries, Pdisk (similar to a file system), Vdisk (similar to a key-value database capable of replicating data), BSController (a tablet managing distributed storage), Blob Depot, and KeyValue tablet. We are constantly improving these and other components and trying to come up with something new.

Open-source development Most of the development is done on GitHub, so you can always see the results of completed tasks. We not only develop and operate a high-load open-source system but also frequently speak at various conferences: HighLoad, CodeFest, Yandex for Developers.

We expect you

• Know classical algorithms and data structures
• Love C++ and have been writing in it for more than three years
• Have experience working with Linux
• Are capable of creating new algorithms and protocols and know how to verify them
• Have developed multithreaded code, can differentiate between loсk-free and wait-free
• Have experience optimizing programs, confidently use Perf and VTune or BpfTrace

Will be a plus

• Know how disks, networks, processors, and memory work
• Have experience developing high-load, distributed, hardware-close services or real-time systems