Назад
Company hidden
10 дней назад

AI Operations Engineer

Формат работы
onsite
Тип работы
fulltime
Грейд
middle/senior
Английский
b2
Страна
CR
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
AI Operations Engineer (AI/Python): Owning and evolving an internal AI Gateway that connects commercial and self-hosted models, with an accent on production engineering, model serving, usage analytics, and AI-ready repositories. Focus on deploying models, reducing costs through routing and context compression, improving CI/CD guardrails, and building secure AI developer tooling across the company.

Location: Prague

Company

hirify.global is a post-Series B network-automation company developing solutions that help enterprises understand and manage their networks, with operations expanding across Europe, the U.S., and beyond.

What you will do

  • Own the internal AI Gateway, including new features, fixes, reliability, and integrations with commercial, third-party-hosted, and self-hosted open-source models.
  • Deploy and serve open-source models, follow model developments, and recommend models and tools to adopt.
  • Aggregate AI usage and spend, build dashboards, and reduce costs through model routing and context compression.
  • Make repositories ready for AI agents by maintaining skills and agent instruction files such as AGENTS.md and CLAUDE.md, improving linters, CI guardrails, and software-factory capabilities.
  • Support engineers in using AI tools effectively, improve the AI developer experience, and contribute to AI features in the hirify.global product.
  • Work with IT, Finance, and Legal on security and data-governance practices.

Requirements

  • Professional experience building and maintaining production software, including sound design, testing, and debugging skills; Python is the main backend language.
  • Comfort working across the stack, including frontend code for dashboards and internal tools, infrastructure, and deployment changes.
  • Experience deploying and operating production services with containers, CI/CD, and observability.
  • Daily hands-on use of AI coding agents such as Claude Code, Cursor, or Codex, with a clear understanding of their strengths and limitations.
  • Practical experience integrating APIs and handling authentication, rate limits, and failures across system boundaries.
  • Ability to follow AI developments and form independent opinions about new models and tools.
  • Clear written and spoken English, including the ability to explain technical choices to people outside engineering.
  • Candidates are considered at L2 or L3 level. At L2, the role owns AI Gateway and agent-readiness work with support on prioritisation and cross-company governance; at L3, the role also leads discussions with IT, Finance, and Legal and sets technical direction.

Nice to have

  • Experience serving LLMs with vLLM, Ollama, TGI, or similar tools, and sizing GPU infrastructure.
  • Experience with LLM gateways or proxies such as LiteLLM or Portkey, and model routing.
  • Experience with usage and cost analytics.
  • Frontend experience with React and TypeScript for internal tools and dashboards.
  • Experience building MCP servers, agent skills, or agent workflows.
  • Awareness of security and data-governance requirements such as SOC 2 and ISO 27001.

Culture & Benefits

  • 25 days of holidays plus 5 flexi-days.
  • Pluxee Flexi Card of up to 16,000 CZK per year and anniversary rewards.
  • Flexible, founder-led startup environment focused on hands-on work, tangible impact, work-life balance, and growth.
  • Collaboration with diverse international teams and regular team and company events.
  • Pet-friendly office at IP Pavlova with snacks, coffee, beer, and refreshments.

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 →