Вакансия из Hirify Global, списка международных tech-компаний Для мэтча и отклика нужен Plus
Пожаловаться
Зарплата и рынок
ЗП не указанаразвернуть
На похожих ролях по рынку
$238к$255кмедиана$325к
50 в выборке
На СНГ рынке: $4.4к/мес ($3.7к - $5.9к)
92
Крутая вакансия
развернуть
Чётко очерченная роль с высокой ответственностью в AI-продуктовой компании, измеримыми целями и работой над быстро развивающимися системами инференса; основной нюанс - задачи могут переключаться между разными нагрузками.
Передовые технологии
Понятные результаты
AI-продукт
?Меняющийся фокус
Дополнительный отдых
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
TL;DR
Member of Technical Staff - GPU Inference Engineer (AI) (vLLM/SGLang): Building and optimizing GPU serving stacks, inference pipelines, and public model integrations for Liquid Foundation Models with an accent on LLM serving performance, latency, throughput, and memory efficiency. Focus on profiling and benchmarking inference engines, improving evaluation and synthetic data generation, and integrating models across vLLM, SGLang, and Hugging Face Transformers.
Location: Hybrid role based in San Francisco or Boston, with remote work available
Company
hirify.global, spun out of MIT CSAIL, builds general-purpose AI systems that run efficiently across data center accelerators and on-device hardware, serving customers in consumer electronics, automotive, life sciences, and financial services.
What you will do
Own and improve the GPU serving stack built on vLLM and SGLang, including configuration, scheduling, and framework changes.
Accelerate large-scale model evaluation and synthetic data generation, and design the inference pipelines supporting these workloads.
Select, profile, benchmark, and tune the inference engine behind the API service for latency and throughput.
Integrate Liquid Foundation Models into vLLM, SGLang, and Hugging Face Transformers, maintaining the public integrations as models and frameworks evolve.
Diagnose throughput and latency issues in collaboration with researchers and serving engineers.
Requirements
Hands-on experience running LLM inference on GPUs with a serving framework such as vLLM or SGLang beyond default configuration.
First-principles understanding of LLM inference, including the transformer forward pass, autoregressive GPU decoding, and latency, throughput, and memory trade-offs.
Solid understanding of GPU fundamentals relevant to inference; writing CUDA kernels is not required.
Experience profiling and benchmarking inference workloads and applying the results.
Strong Python and PyTorch skills, with the ability to read and modify large open-source codebases.
Nice to have
Merged contributions to vLLM, SGLang, or FlashInfer.
Lower-level GPU optimization experience, such as CUDA or Triton kernels.
Experience building high-throughput batch inference for evaluation or synthetic data generation.
Culture & Benefits
High-ownership work focused on solving inference problems that standard serving stacks do not handle directly.
Competitive base salary plus equity; final offers reflect skills, experience, and location.
Unlimited PTO, company-wide Refill Days, and shared long weekends.
Comprehensive health coverage, with details varying by location.
Paid leave for pregnancy, parenting, medical needs, and family caregiving.
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 →