Роль с высоким уровнем ответственности в быстрорастущей AI-инфраструктурной компании с четко определенными задачами и современным высокопроизводительным стеком.
Трендовый домен
?Современный стек
Четкие обязанности
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
TL;DR
Principal Engineer, IaaS (AI Cloud Infrastructure): Building and scaling multi-tenant IaaS control and data planes across compute, high-performance networking, and storage with an accent on performance, reliability, and efficiency for large AI workloads. Focus on designing distributed infrastructure, optimizing low-level data paths with RDMA and GPU networking technologies, and developing production systems in Go, C, C++, or Rust.
Location: Chicago, Illinois, United States
Base salary: $206,000–$275,000 per year, plus discretionary bonus, equity awards, and benefits subject to eligibility.
Company
CoreWeave provides cloud infrastructure, tools, and services for AI labs, startups, enterprises, and open-source communities.
What you will do
Set technical direction and architecture standards for the IaaS organization.
Design and implement scalable, multi-tenant control and data planes across compute, networking, and storage.
Build GPU and bare-metal provisioning, high-performance network fabrics, and exabyte-scale object, block, and distributed file storage.
Optimize compute, network, and storage data paths using technologies such as RDMA, GPU Direct, RoCE, InfiniBand, SPDK, virtualization, and Kubernetes.
Improve platform reliability, durability, scalability, observability, and operational performance through telemetry, metrics, and dashboards.
Write production systems code, mentor engineers, and collaborate with compute, networking, storage, platform, product, and operations teams.
Requirements
8–10+ years of experience building infrastructure, distributed systems, or cloud services in production.
Hands-on experience operating large-scale systems in at least one core domain: compute and virtualization, high-performance networking, or object, block, or distributed filesystem storage.
Proficiency in a systems programming language such as Go, C, C++, or Rust.
Strong understanding of cloud-native infrastructure, Kubernetes, and scalable system architecture.
Strong debugging and problem-solving skills in distributed, high-performance environments, including identifying bottlenecks and failure modes.
Ability to communicate complex architectural trade-offs to engineers, executives, and customers and influence decisions across teams.
Bachelor’s, Master’s, or PhD in computer science, electrical engineering, or a related field, or equivalent practical experience.
Nice to have
Experience with RDMA, GPU Direct, RoCE, InfiniBand, SPDK, or large-scale provisioning, scheduling, and orchestration systems.
Familiarity with infrastructure hardware trade-offs, including HDD versus SSD storage tiers and network fabric selection.
Experience operating AI/ML or other GPU-accelerated, high-throughput infrastructure.
Publications or talks on large-scale distributed systems or infrastructure.
Culture & Benefits
Full-time employment with US-based benefits, including fully paid medical, dental, and vision insurance.
Life, disability, mental wellness, family-forming, parental leave, childcare, flexible spending, and health savings benefits.
401(k) with employer match, tuition reimbursement, employee stock purchase program, flexible PTO, discretionary bonus, and equity awards.
Office and data center environment with catered lunch and a culture focused on innovative disruption.
Access to technology and source code is subject to US export control requirements.
Hiring process
Export-control information may be requested during recruitment to assess applicable US requirements.
Reasonable accommodations are available for qualified candidates with disabilities upon request.
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 →