Accountabilities

• Develop and optimize YDB components to take full advantage of modern hardware, including QLC NVMe drives, high-bandwidth network adapters, and DPUs.
• Improve performance across widely used storage technologies, including HDDs and TLC NVMe devices.
• Reengineer system components using more efficient algorithms to address complex scalability, performance, and reliability challenges.
• Design and implement high-performance, low-latency software components for heavily loaded production systems.
• Analyze system behavior and identify performance bottlenecks using profiling, debugging, and diagnostic techniques.
• Work on distributed storage and database infrastructure supporting demanding AI and cloud workloads.
• Contribute to the reliability, availability, and durability of production-grade distributed systems.
• Participate actively in incident investigation and resolution, helping identify root causes and implement lasting improvements.
• Collaborate with engineers across infrastructure, storage, networking, and AI-focused teams to deliver robust technical solutions.
• Contribute to an open-source codebase and participate in technical discussions, design decisions, and continuous improvement.
• Take ownership of engineering challenges from investigation and design through implementation, optimization, testing, and production deployment.
• 
  

Requirements

• 5+ years of professional experience developing in C or C++ for highly loaded, performance-critical systems.
• Strong understanding of low-level systems concepts, including CPU caches, modern CPU atomic operations, and NUMA architectures.
• Proven experience developing high-performance and low-latency software components.
• Ability to investigate complex production issues using core dumps, flame graphs, sanitized builds, and related debugging techniques.
• Strong systems programming and performance optimization skills, with an analytical approach to diagnosing bottlenecks.
• Experience with on-disk data structures such as LSM trees or B+ trees is a strong advantage.
• Hands-on experience with tools such as perf, VTune, bpftrace, or gdb is desirable.
• Knowledge of storage algorithms, including erasure coding and checksumming, is a plus.
• Understanding of storage device internals, particularly NVMe and HDD technologies.
• Familiarity with Linux kernel development and technologies such as SPDK, DPDK, libaio, or io_uring is advantageous.
• Knowledge of networking concepts and protocols including IP, TCP, UDP, and DNS; experience with InfiniBand, RoCE, or RDMA is a plus.
• Experience with Kubernetes and Grafana is desirable.
• Experience designing production-grade distributed storage components and understanding availability and durability calculations is highly valued.
• Demonstrated ability to troubleshoot complex systems, participate in incident resolution, and work effectively in collaborative engineering environments.
• Strong ownership, problem-solving skills, and enthusiasm for tackling technically challenging infrastructure problems.
• Willingness to participate in coding interviews as part of the recruitment process.