1 year ago
Remote, WorldwideSenior
Responsibilities
- Design and evolve production Ceph clusters, including hardware design, network requirements, configuration, tuning, and client operations.
- Create efficient APIs using systems and kernel features to enable safe, high-performance live migration of stateful workloads between hosts.
- Design and build storage API and orchestration services using Go, gRPC, ScyllaDB, and Temporal.
- Write Engineering Requirement Documents covering ideas, implementation tasks, and success monitoring.
- Build reusable storage primitives for customer applications, internal services, streaming image pulls, movable build caches, and other platform features.
- Participate in technical interviews, solution reviews, and team collaboration.
Requirements
- Experience architecting and implementing fault-tolerant, resilient, and scalable distributed systems.
- Production experience with distributed block device systems such as Ceph, or a strong understanding of network storage cluster design from first principles.
- Understanding and experience with current-generation filesystems including Ext4, ZFS, and BTRFS; EROFS and bcachefs experience is a bonus.
- Ability to reason about system longevity, implement solutions, monitor error boundaries, and document operational requirements.
- Strong prioritization and direction-setting skills in an ambiguous early-stage startup.
- Grit to develop, scale, and eventually replace systems when needed.
- Strong communication skills for explaining solutions and working with others.
- Ability to complete and discuss a pre-interview storage engine design project.
Benefits
- Full health benefits including dependents
- Equity grants
- Equipment stipend
- Globally distributed work environment
- Few meetings and high autonomy
- High-ownership culture with opportunities for growth
- Novel, high-leverage startup problems
About Railway
Railway is a developer-focused Platform as a Service that lets teams deploy containerized applications, APIs, and managed databases without managing infrastructure. The company offers a usage-based, self-serve cloud platform with workflows for builds, environments, logs, and secrets, aimed at solo builders through startups. Founded in 2020 and headquartered in San Francisco, it is privately held.