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
$180к$218кмедиана$252к
48 в выборке
На СНГ рынке: $4.7к/мес ($4.6к - $8.6к)
91
Крутая вакансия
развернуть
Чётко очерченная системная роль на стыке крупномасштабных вычислений и робототехнического ИИ даёт серьёзную техническую глубину, хотя работа в офисе в Сан-Франциско ограничивает гибкость.
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
TL;DR
ML Infra Engineer (AI): Building intelligent scheduling, placement, and orchestration systems for large-scale GPU and TPU training clusters with an accent on resource allocation, multi-cluster control planes, and fault-tolerant distributed systems. Focus on optimizing accelerator utilization, implementing priority and preemption policies, and extending compute infrastructure to inference workloads and robot deployment.
Location: San Francisco, United States; Onsite
Company
hirify.global develops general-purpose AI, foundation models, and learning algorithms for robots and physically actuated devices.
What you will do
Design and build multi-tenant scheduling and placement systems for training jobs across heterogeneous GPU and TPU clusters.
Develop multi-cluster orchestration and control-plane capabilities for job lifecycle management, migration, failover, and rescheduling across on-premises and cloud environments.
Optimize accelerator utilization through quota management, priority tiers, preemption, queueing, and fairness policies.
Implement fault detection, automatic recovery, health checking, node management, and scaling for long-running multi-node jobs across thousands of accelerators.
Extend scheduling and orchestration to inference workloads and model deployment on physical robots.
Build dashboards, alerting, SLOs, and debugging tools to improve observability, researcher experience, and cluster reliability.
Requirements
Strong software engineering fundamentals.
Experience building or operating job scheduling or resource management systems at scale.
Experience with large-scale compute clusters using GPUs and/or TPUs.
Familiarity with schedulers and orchestration systems such as SLURM, Kubernetes, GKE, K3S, or internal equivalents.
Ability to reason about resource allocation, bin-packing, priority scheduling, and multi-tenancy.
Understanding of long-running, multi-node ML training workloads, including stragglers and topology dependencies.
Willingness to own systems end-to-end, from design through operation, and work closely with researchers to unblock projects.
Nice to have
Experience building multi-cluster or federated scheduling systems.
Experience with TPU infrastructure, including GCP TPU slices, Multislice, or GKE.
Background with Borg, YARN, Mesos, or custom cluster resource managers.
Linux systems engineering, networking, and infrastructure-as-code experience.
Experience with NCCL, collective communication, and topology-aware placement.
Experience with capacity planning and cloud cost optimization at scale.
Familiarity with JAX, PyTorch, or similar ML frameworks at the runtime or systems level.
Culture & Benefits
Full-time, onsite work in San Francisco.
Close collaboration with ML infrastructure, data platform, and research teams.
Employment consideration is subject to the San Francisco Fair Chance Ordinance; qualified applicants with arrest and conviction records will be considered.
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 →