Вакансия из Hirify Global, списка международных tech-компаний Для мэтча и отклика нужен Plus
Пожаловаться
Зарплата и рынок
ЗП не указанаразвернуть
На похожих ролях по рынку
$120к$165кмедиана$204к
49 в выборке
На СНГ рынке: $2.5к/мес ($1.6к - $3.6к)
87
Хорошая вакансия
развернуть
Это четко очерченная сетевая роль на стыке инфраструктуры дата-центров и передового ИИ, с ответственностью от проектирования до эксплуатации. Темп стартапа и обязательная работа из офиса подойдут не всем.
Только офис
Полный цикл задач
Современные технологии
Инфраструктура ИИ
?Динамичная среда
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
TL;DR
MTS, Supercomputing Networking (AI): Designing and operating RDMA-based data center networks for AI inference systems with an accent on network automation, telemetry, and production reliability. Focus on diagnosing bandwidth and latency issues, building safe rollout and recovery workflows, and qualifying network hardware under failure conditions.
Location: San Jose, United States; fully in-person at the San Jose office. Relocation support is available for candidates moving to San Jose.
Company
hirify.global builds custom chips, racks, software, and manufacturing systems for frontier AI inference, with a focus on throughput and latency.
What you will do
Design and bring up data center networks connecting inference nodes across racks, including RDMA fabrics.
Automate network provisioning, configuration, validation, staged rollouts, rollback, and recovery.
Build telemetry and diagnostics for network health, congestion, failures, and issues across hosts, NICs, and switches.
Investigate bandwidth, latency, and reliability problems across network configuration, Linux, drivers, and firmware.
Develop qualification, burn-in, and failure tests for NICs, switches, and network configurations.
Partner with hardware, firmware, system software, fleet, and deployment teams and work with vendors to qualify equipment and firmware.
Requirements
Experience building and operating production data center networks, including deployment, troubleshooting, and reliability ownership.
Hands-on RDMA networking experience, including RoCE, with an understanding of fabric configuration and congestion effects on performance.
Strong fundamentals in Ethernet, IP routing, switching, and network security.
Experience writing production software or automation for network provisioning, configuration management, monitoring, or remediation.
Proficiency in Python, Go, C++, Rust, or a similar language.
Strong Linux fundamentals and the ability to debug hosts, NICs, and switches using logs, counters, packet captures, and other diagnostics.
Experience owning technical work from design through deployment and production support.
Comfort working across software, hardware, and operations teams.
Nice to have
WAN or backbone networking experience, including routing, traffic engineering, and data center connectivity.
Experience operating CDN, cloud, or other large-scale network infrastructure.
Experience with GPU, TPU, or other accelerator clusters.
Familiarity with SONiC, gNMI, OpenConfig, or similar switch operating systems and management interfaces.
Experience tuning RDMA congestion control and diagnosing packet loss, queueing, and uneven traffic distribution.
Experience building network telemetry pipelines and operating failure testing, failover, and recovery at scale.
Culture & Benefits
Full-time, fully in-person work in San Jose, with technical staff contributing across engineering and research disciplines.
Medical, dental, and vision coverage, including a $500 monthly credit for waiving medical benefits.
$2.5k monthly housing subsidy for those living within walking distance of the office.
Relocation support for moves to San Jose, wellness benefits, and daily office lunch and dinner.
Unlimited compute budget subject to ROI justification.
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 →