OpenAI13 дней назад

Инженер-программист, рантайм AI-акселератора

Зарплата не указана
РЫНОК
26 250медиана по профессии
AI Researcher · 34 вакансий с указанной зарплатой
5 600половина предложений: 16 690–59 425143 300
Работодатель не указал зарплату — сравните с рынком сами.
Полная занятостьУдалёнка

Обязанности

  • 01Design and implement the low-level device runtime for OpenAI custom silicon
  • 02Build kernel-launch scheduling, command submission, queueing, dependency tracking, and completion handling
  • 03Manage device memory spaces, allocation, virtual-to-physical mappings, data movement, and lifetime across concurrent workloads
  • 04Implement synchronization primitives, events, barriers, streams, and ordering guarantees that are correct and efficient
  • 05Define clean interfaces between the runtime, drivers, firmware, compiler-generated code, kernels, and higher-level execution systems
  • 06Use event-based, cycle-accurate simulators to develop, validate, debug, and performance-tune runtime behavior before and after silicon availability
  • 07Diagnose concurrency, memory-ordering, deadlock, race, correctness, and performance issues across software and hardware boundaries
  • 08Build tests, tracing, profiling, observability, and reproducible workloads for runtime correctness and performance
  • 09Partner with architecture and silicon teams to turn workload and simulator insights into hardware-software interface improvements

Требования

  • 01Have strong low-level systems programming experience in C, C++, Rust, or comparable environments
  • 02Have built runtimes, drivers, firmware, operating-system components, accelerator software, or adjacent systems infrastructure
  • 03Understand concurrency, synchronization, asynchronous execution, queues, events, and memory-ordering semantics
  • 04Understand memory management, address spaces, DMA, caching, coherency, and hardware-software interfaces
  • 05Have hands-on experience with event-based, cycle-accurate simulators or closely related architectural and performance models
  • 06Can reason quantitatively about scheduling, latency, throughput, utilization, contention, and resource tradeoffs
  • 07Are skilled at debugging failures that span software abstractions, device interfaces, and hardware behavior
  • 08Work effectively across compiler, kernel, architecture, verification, firmware, and silicon teams
  • 09To comply with U.S. export control laws and regulations, candidates for this role may need to meet certain legal status requirements as provided in those laws and regulations

Условия

  • 01We are committed to providing reasonable accommodations to applicants with disabilities