Назад
Company hidden
6 дней назад

Head of Inference (AI)

Формат работы
onsite
Тип работы
fulltime
Грейд
head
Английский
b2
Страна
US
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
Head of Inference (AI): Building and owning the enterprise AI inference pillar, from procuring bare-metal GPU capacity and deploying open-weight models to serving production endpoints, with an accent on inference optimization, heterogeneous accelerators, and cost-per-token efficiency. Focus on managing KV caches, continuous batching, quantization, autoscaling, and production infrastructure while building the inference team.

Location: Sunnyvale, California, US

Company

hirify.global builds CanyonOS, an AI infrastructure platform focused on design, deployment, optimization, and inference for enterprise AI workflows.

What you will do

  • Own the inference pillar and the complete path from GPU capacity procurement to production model serving.
  • Procure bare-metal GPU capacity from neoclouds and colocation facilities, including provider evaluation, negotiations, delivery, and production readiness.
  • Deploy open-weight models and serve the production endpoints used by customers.
  • Make capacity, provider, hardware mix, and serving stack decisions.
  • Work directly with the CEO and Chief Architect on AI infrastructure.
  • Hire, mentor, and build the inference team as the function grows.

Requirements

  • More than once within the last couple of years, independently completed the full cycle of procuring bare-metal GPU capacity, deploying open-weight models, and serving production endpoints.
  • Hands-on experience with vLLM, SGLang, TensorRT-LLM, or NVIDIA Dynamo.
  • Strong understanding of KV-cache management, continuous batching, paged attention, speculative decoding, and quantization, including their workload-specific trade-offs.
  • Experience owning and improving a cost-per-token metric.
  • Strong systems and operations skills across Kubernetes, containers, networking, storage, autoscaling, and observability.
  • Experience with heterogeneous accelerators: NVIDIA plus at least one of AMD, Groq, Cerebras, Intel, or TPU.
  • Ability to hire, mentor, and build a team.
  • Self-driven approach and comfort with ambiguous, early-stage work.

Nice to have

  • Experience inside an inference provider such as Baseten, Fireworks, Lightning, Together, or Modal.
  • Existing relationships with neoclouds or colocation providers.
  • Experience with agentic workloads involving many model calls per request.
  • Published or open-source work in inference optimization.

Culture & Benefits

  • Full-time founding-team role owning a core company pillar.
  • Direct collaboration with the CEO and Chief Architect.
  • Ownership of architecture, providers, hardware mix, and hiring.
  • Work in an early-stage environment with capital available for capacity and infrastructure decisions.
  • Start from CanyonOS's measured advantages in harness scaling and LLM serving.

Hiring process

  • Send a resume and a short note explaining what excites you about the role.
  • Include Head of Inference in the subject line and apply at careers@canyoncode.ai.

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 →