Назад
обновлено 2 месяца назад

Gpu Kernel Engineer (AI)

180 000 - 360 000$
Формат работы
hybrid
Тип работы
fulltime
Грейд
middle/senior
Английский
b2
Страна
US/Canada
Вакансия из списка Hirify.GlobalВакансия из Hirify Global, списка международных tech-компаний
Для мэтча и отклика нужен Plus

Мэтч & Сопровод

Для мэтча с этой вакансией нужен Plus

Описание вакансии

Текст:
/
TL;DR
Gpu Kernel Engineer (AI): Develop and optimize high-performance GPU kernels for machine learning workloads with an accent on CUDA programming, kernel optimization, and advanced GPU features. Focus on designing efficient GPU operations, quantization techniques, and performance profiling to accelerate AI model inference.

Location: Hybrid in San Francisco, Montreal, New York, Seattle, or Toronto

Salary: $180,000–$360,000 plus equity

Company

Baseten provides inference infrastructure, applied AI research, and developer tooling for deploying machine learning models in production.

What you will do

  • Design and implement high-performance GPU kernels for matrix multiplication, attention mechanisms, and mixture-of-experts routing.
  • Write and optimize CUDA, PTX assembly, and architecture-specific GPU code.
  • Apply memory coalescing, warp-level programming, tensor core acceleration, and compute/memory overlap techniques.
  • Implement quantization, including FP8 and FP4, sparsity, and communication-overlap features.
  • Identify performance bottlenecks with Nsight Systems, Nsight Compute, and Torch Profiler.
  • Collaborate with research teams and contribute to internal and open-source GPU libraries.

Requirements

  • Strong understanding of GPU architecture, memory hierarchy, thread/block/grid organization, synchronization, and race-condition mitigation.
  • Proficiency in C++ and GPU performance profiling tools.
  • Knowledge of the CUDA C++ API, memory access patterns, and bandwidth optimization.
  • Understanding of numerical precision, quantization strategies, tensor cores, and asynchronous GPU operations.

Nice to have

  • Experience with Transformer models and attention optimization, including Flash Attention.
  • Familiarity with Cutlass, Triton, Thrust, or CUB.
  • Experience with GEMM tuning and distributed or multi-GPU computing.
  • Open-source GPU contributions, research publications, or conference presentations on GPU performance.

Culture & Benefits

  • Competitive compensation with meaningful equity.
  • Flexible PTO and a company-wide winter break.
  • Paid parental leave and a fertility and family-building stipend.
  • U.S. employees receive full medical, dental, and vision insurance coverage for themselves and dependents.
  • U.S. employees have access to a company-facilitated 401(k).
  • Exposure to a range of machine learning startups and production AI systems.

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 →