Чётко очерченная инфраструктурная роль уровня Staff в ведущей AI-компании даёт редкий опыт работы с системами петабайтного масштаба и напрямую связана с обучением передовых моделей.
Современный стек
Инфраструктура AI
Сильный соцпакет
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
TL;DR
Staff Software Engineer (AI/Data Infrastructure): Architecting and operating a unified distributed storage layer for petabyte-scale model training and evaluation with an accent on storage architecture, Kubernetes deployment, and high-throughput data movement. Focus on serving training data and model checkpoints across clusters, reducing GPU idle time, and solving networking, I/O, consistency, latency, and durability challenges.
Location: New York, United States. Cohere is remote-friendly, with offices in Toronto, London, New York City, San Francisco, Montreal, Paris, Berlin, and Seoul.
Salary: United States — California, New York, and Washington: base salary $275K–$335K plus equity; all other US states: $245K–$300K plus equity; Canada: CA$350K–CA$420K plus equity.
Company
Cohere is a security-first enterprise AI company building foundation models and end-to-end products for real-world business problems.
What you will do
Design, architect, and operate the distributed storage system serving model training and evaluation.
Lead deployment across multiple Kubernetes clusters at petabyte scale and establish organizational standards.
Collaborate with research and training-infrastructure teams to translate data-access patterns into throughput, latency, and durability requirements.
Solve networking, I/O, consistency, and cross-region or cross-backend data-movement challenges for large datasets and model checkpoints.
Requirements
Strong storage fundamentals covering replication, consistency, caching, and data lifecycle management, with architecture and design experience.
Strong coding ability in Python and Go, with the ability to lead implementations and mentor others in either language while contributing to the other.
Experience running stateful systems on Kubernetes, including Persistent Volumes, CSI drivers, and StatefulSets, plus the ability to design and execute operational strategies.
Hands-on experience with cloud object storage such as S3 and POSIX-style filesystems.
Nice to have
Experience with parallel or HPC filesystems such as Weka, VAST, or Lustre applied at scale.
Familiarity with data-loading and checkpointing patterns used in large-scale model training.
Interest in how large models are trained and evaluated, including the trajectory and evaluation data generated by those runs.
Culture & Benefits
Full-time employment with remote-friendly and office-based work options; remote employees receive a coworking benefit, home-office stipend, office-travel budget, and annual company offsite.
Full health and dental benefits, mental-health budget, and retirement contributions through RRSP matching, 401(k), or a pension scheme.
Six weeks of paid vacation and up to six months of 100% parental-leave top-up for either parent.
Weekly lunch stipend of $75/£75 or local equivalent, plus annual enrichment and education benefits for culture, wellness, conferences, courses, and coaching.
Inclusive work environment with accommodations available during the recruitment process.
Hiring process
Applicants may be screened and assessed using AI-enabled tools; accommodations can be requested through the provided form.
Legitimate roles and communications are available only through Cohere’s official careers channels and @cohere.com or @cw.cohere email aliases; no payment or third-party services are requested.
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 →