В OpenAI это чётко очерченная технически сложная роль на переднем крае безопасности ИИ, с ответственностью за проекты целиком и тесной работой с исследователями; основной нюанс - гибридный формат с привязкой к офису в Лондоне.
?Привязка к Лондону
Безопасность ИИ
Полная ответственность
Чёткие задачи
Современный ИИ-стек
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
TL;DR
Software Engineer, Guardian (AI): Building production systems that evaluate AI-agent behavior, detect risk, and enable safe recovery or escalation with an accent on real-time oversight, user intent, privacy, and operational reliability. Focus on translating alignment research into scalable product capabilities, designing adversarial evaluations, and integrating safety systems across agent experiences.
Location: London, UK; hybrid, based in the London office
Company
OpenAI is an AI research and deployment company building general-purpose AI systems with safety and human needs at their core.
What you will do
Design, build, and operate core components of Auto-review and the Guardian platform, balancing safety, product utility, latency, and operational reliability.
Build real-time production systems that evaluate agent actions, detect potentially harmful behavior, and support safe recovery or escalation.
Improve oversight systems’ understanding of user intent, environmental context, permissions, and personalization signals while protecting privacy.
Translate alignment research, prototypes, evaluations, and monitoring techniques into scalable product capabilities.
Integrate Guardian capabilities into agent experiences with Product, Security, Infrastructure, and engineering teams across OpenAI.
Develop evaluations, adversarial tests, and monitoring; own projects end to end and investigate failures to improve reliability.
Requirements
Experience building and shipping complex production software with ownership across design, implementation, deployment, and operation.
Experience building with AI agents or large language models in a core product, startup, or technically serious independent project, with an understanding of agentic-system challenges.
Strong software engineering fundamentals and the ability to design reliable backend services, APIs, and infrastructure.
Sound product-engineering judgment when balancing safety, usability, speed, and reliability.
Ability to work closely with researchers, translate open-ended ideas into practical systems, and use experiments and evaluations to guide decisions.
Comfort with ambiguity, initiative, and clear communication across technical disciplines.
Commitment to building AI systems that are useful, trustworthy, and safe.
Culture & Benefits
Hybrid work based in the London office within a concentrated, highly collaborative engineering group.
Cross-functional collaboration across engineering, alignment research, product, infrastructure, and security.
Equal opportunity employment and reasonable accommodations for applicants with disabilities.
Background checks are administered in accordance with applicable law.
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 →