15 часов назад
Senior Software Engineer (AI/HPC Middleware)
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Описание вакансии
Текст:
TL;DR
Senior Software Engineer (AI/HPC Middleware): Developing and optimizing HPC and AI communication middleware, including MPI, SHMEM, and collective communication libraries, over high-performance interconnects with an accent on low-latency communication, transport integration, and upstream open-source contributions. Focus on profiling and tuning distributed communication paths, validating cluster-scale performance, and solving customer issues across drivers, switches, systems, and application frameworks.
Location: Remote for employees residing within the United States; the position is associated with New York, New York.
Company
develops high-performance scale-out networking solutions for AI and HPC datacenters, integrating silicon, software, and system technologies for GPU, CPU, and accelerator clusters.
What you will do
- Design, develop, and optimize HPC middleware such as MPI and SHMEM, along with AI collective communication libraries including NCCL and RCCL.
- Integrate middleware capabilities with transport and provider layers such as OFI/libfabric, UCX, and verbs-style interfaces.
- Prepare and submit upstream contributions to MPI, SHMEM, collective communication, and related open-source projects.
- Collaborate with kernel, driver, switch, systems, and application teams to deliver end-to-end performance.
- Analyze performance traces, reproduce customer issues, and translate findings into fixes for real-world workloads.
Requirements
- 6+ years of systems programming experience in C/C++ on Linux and a bachelor's degree in computer science, engineering, or a related field, or equivalent experience.
- Hands-on experience with HPC middleware internals such as Open MPI, MPICH, MVAPICH, or SHMEM, or with AI collective communications including NCCL, RCCL, CUDA, or ROCm.
- Experience delivering low-latency and high-throughput communication paths and diagnosing performance issues with profiling and tracing tools.
- Working knowledge of OFI/libfabric, UCX, verbs-style concepts, RDMA, and performance tuning.
- Experience contributing to open-source projects.
- Applicants must reside within the United States.
Nice to have
- Experience developing or maintaining libfabric providers.
- Familiarity with Ultra Ethernet specifications, RoCEv2, congestion control, or Ethernet-based RDMA.
- Experience with cluster-scale benchmarking, profiling, and optimization.
- Background with HPC fabrics such as Omni-Path.
Culture & Benefits
- Flexible remote work environment within the United States.
- Competitive compensation including base pay, equity, cash, and performance-based incentives.
- Medical, dental, vision, disability, life, dependent care, accidental injury, and pet insurance.
- Paid holidays, sick time, bonding leave, pregnancy disability leave, and Open Time Off for eligible full-time exempt employees.
- 401(k) plan with company match.
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 →
Похожие вакансии
20 часов назад
Staff Software Engineer, Systems (AI)
136 500 - 260 500$
6 часов назад
Senior Middleware Software Development Engineer (HPC)
128 880 - 245 160$
2 дня назад
New Graduate Software Engineer (AI)
2 часа назад
Senior Software Engineer (Java)
Nebius
7 дней назад
Senior Software Engineer (C++)
ClickHouse
7 дней назад