Роль с высокой ответственностью в известной продуктовой компании: задача четко описана, инфраструктурные технологии актуальны, а удаленный формат и возможности обучения привлекательны. При этом от вас ждут серьезного senior-уровня и участия в дежурствах.
?Высокая ответственность
Актуальные технологии
Продуманные условия
Влияние на продукт
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
TL;DR
Storage Platform Engineer (NVMe-oF): Building a software-defined, NVMe-over-Fabrics storage platform for Supabase Postgres with an accent on networked block storage, durability, performance, and multi-cloud portability. Focus on integrating Kubernetes CSI infrastructure, tuning the full Postgres I/O path, designing replication or erasure coding, and enabling snapshots and copy-on-write clones.
Location: Remote, global; work from anywhere
Company
Supabase is an open-source-first platform running millions of Postgres databases with a globally distributed team.
What you will do
Lead the architecture and build-versus-buy decisions for a software-defined NVMe-over-Fabrics storage platform.
Integrate networked block storage with the Kubernetes-based compute platform through CSI drivers and custom controllers for provisioning, attachment, resizing, and migration.
Profile and tune the full Postgres I/O path, from fsync, WAL, and checkpoints through the Linux block layer, NVMe/TCP, and the network.
Design durability, availability, replication or erasure coding, failure handling, failover, data-integrity verification, I/O isolation, and QoS.
Build SLOs, benchmarking, observability, block-level snapshots, and copy-on-write clones for database branching, while keeping the platform portable across clouds and bare metal.
Write design documents, mentor engineers, collaborate with Compute, Branching, Postgres, and Multicloud teams, and participate in on-call, incident response, and systemic issue resolution.
Requirements
7+ years of software engineering experience, including significant experience building and operating storage systems in production.
Experience building or operating distributed or software-defined block storage such as Ceph RBD, Lightbits, Simplyblock, Portworx, LINSTOR/DRBD, or an in-house system.
Hands-on experience with NVMe/TCP or NVMe-oF, including targets, initiators, multipath, and failover behavior.
Deep knowledge of Linux storage internals, including the block layer, io_uring, page cache, O_DIRECT and fsync semantics, filesystems, and device-mapper.
Strong systems programming skills in Go plus C, C++, or Rust.
Rigorous performance measurement with fio, pgbench, eBPF, and perf, including explaining latency components.
Understanding of durability tradeoffs between replication, erasure coding, and consistency under failure.
Experience leading technical direction for complex systems through design documents and tradeoff decisions.
Experience with Kubernetes controllers and/or CSI drivers and being on call for large production systems.
Clear written communication and ability to work in an async, globally distributed environment.
Postgres internals knowledge, including WAL, buffer manager, checkpointer, or familiarity with OrioleDB.
Experience with AWS networking and instance storage at scale, including ENA, placement, and bandwidth limits.
Experience running storage on bare metal or in colocation, or contributing to open-source storage projects or the Linux kernel storage stack.
Culture & Benefits
Fully remote and globally available; there are no Supabase offices, with a WeWork membership or coworking allowance available worldwide.
Asynchronous, flexible work with trust to manage working time.
ESOP equity ownership for every team member.
Tech allowance, annual education allowance, and health insurance covering 100% for employees and 80% for dependents.
Annual company off-site and a remote, open-source-first environment with a team distributed across 60+ countries.
Будьте осторожны
Если работодатель просит войти в их систему, используя iCloud/Google, прислать код/пароль, запустить код/ПО, не делайте этого. Это мошенники. Обязательно жмите «Пожаловаться» или пишите в поддержку. Подробнее в гайде →