aisiemllmdata pipelinessecurity engineeringcloud securitysoarred teamingidentity and access management
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
$147к$178кмедиана$225к
50 в выборке
На СНГ рынке: $4.4к/мес ($3.6к - $5.5к)
86
Крутая вакансия
развернуть
Чёткая и значимая роль на стыке AI-разработки и безопасности, с хорошими условиями и продуктовой средой; при этом широкий охват предполагает ответственность сразу за несколько направлений защиты.
?Широкая зона ответственности
Сильный пакет льгот
Безопасность AI
?Офис три дня
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
TL;DR
Staff AI Engineer (Security/Infrastructure) (AI/Security): Building secure AI agent infrastructure, identity and access controls, detection and response automation, and security evaluation pipelines with an accent on agent security, cloud and application security, and production AI engineering. Focus on threat-modeling agent systems, enforcing least-privilege access, automating SIEM/SOAR responses, and proving resilience through red-team exercises.
Location: London, UK. Hybrid working with 3 days per week in the office; employees may work from anywhere for up to 10 days per year.
Company
hirify.global is an AI adoption and upskilling platform providing apprenticeships and adaptive learning technology to help companies develop AI, data, and technology capabilities.
What you will do
Build security into the path to production using policy as code, review agents, baseline checks, and risk-based deployment gates.
Threat-model AI agent systems and design controls for prompt injection, tool misuse, context and retrieval data leakage, sandboxing, scoped permissions, and human approval.
Build identity and access management for agents, including named ownership, least-privilege access, revocation, and activity records.
Develop SIEM and SOAR detections, automated alert triage, context gathering, and bounded, reversible responses.
Secure workplace devices, tools, accounts, and joiner-mover-leaver access through automated controls.
Build security evaluation pipelines and red-team exercises, contribute to architecture decisions, and influence technical direction.
Requirements
Experience building security into systems across application or cloud security, identity and access, and detection and response.
Understanding of attacks against AI systems, including prompt injection, tool misuse, and data leakage through context, with experience designing preventative controls.
Experience shipping production AI systems serving real users at scale, including context management, model selection and routing, cost engineering, tool use, agent augmentation, and evaluation.
Full-stack delivery experience across LLM integration, backend services, data pipelines, and enough frontend development to ship end to end.
Product mindset with the ability to identify user problems, understand requirements, test and iterate quickly, and execute under ambiguity.
Ability to explain technical concepts clearly to engineers and business colleagues and proactively bring stakeholders along.
Fluency in AI-assisted development and ability to mentor others in using it effectively.
Culture & Benefits
27 days of holiday including winter shutdown days and bank holidays, plus 5 additional days off for life events, volunteering, and summer shutdown.
Private medical insurance, medical cashback, therapy, gym access, life insurance, workplace pension, Cycle to Work, and Tech Scheme.
Permanent employees receive share options; enhanced maternity, paternity, adoption, and shared parental leave are available.
Quarterly team socials, company-wide events, weekly office lunch, snacks, and drinks.
Successful applicants must complete at least a Basic DBS check; Enhanced DBS and additional checks may apply to roles involving Regulated Activity.
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 →