Remote job
Principal Storage Software Engineer
Job details
About this role
Role overview Set the technical direction for distributed storage services and lead their design, implementation, and operation. The platform supports object, block, and file storage, with demanding needs around performance, durability, availability, and data integrity. You’ll work across storage, compute, networking, and Kubernetes integrations, including workloads for AI, high-performance computing, and other data-intensive use cases.
Responsibilities - Design storage services across control and data planes, and connect them with existing platform capabilities. - Translate customer and product needs into architectures, APIs, and technical designs with product partners. - Define credible service levels and own performance, consistency, durability, and availability characteristics. - Benchmark and tune throughput, latency, and IOPS for training, inference, and other demanding workloads. - Integrate storage with compute, Kubernetes CSI, networking, IAM, observability, and developer tooling. - Design replication, snapshots, recovery, data protection, and safe upgrades; partner with operations teams on capacity planning, incidents, and improvements.
Requirements - Extensive experience building and operating large-scale distributed storage systems. - Strong understanding of storage architecture, including how hardware and design choices affect workload performance. - Ability to make long-term architectural decisions where data integrity and availability are critical. - Strong engineering practices, including writing and reviewing code for correctness, performance, and maintainability. - Technical leadership across teams, with experience mentoring engineers and setting standards. - Clear written and verbal communication, including customer discussions on complex technical problems.
Nice to have - Familiarity with storage systems such as WEKA, VAST, DDN, Lustre, Ceph, MinIO, or Pure Storage. - Understanding of AI and HPC storage patterns, including training data pipelines, checkpointing, and parallel file systems. - Experience with durability and consistency models, failure recovery, observability, IAM, encryption, or auditability.
Benefits and work setup Includes 25 days of annual leave, dedicated learning time, private medical coverage, a cycle-to-work scheme, gym and wellbeing app access, company share participation, and enhanced parental pay and leave.