В OpenAI четко описана исследовательская роль на стыке передового ИИ и безопасности: от экспериментов до внедрения защитных мер. Главный практический нюанс - гибридный формат в Сан-Франциско.
Выше рынка
Передовой ИИ
Четкая зона ответственности
Влияние на продукт
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
TL;DR
Researcher, Frontier Robustness (AI): Developing and testing methods to secure frontier AI systems, including containment, monitoring, adversarial evaluation, and layered defenses with an accent on sandbox security, model behavior, and threat modeling. Focus on designing reproducible experiments, discovering cross-boundary vulnerabilities, building automated red-team evaluations, and translating findings into deployed safeguards.
Location: San Francisco, United States; hybrid workplace
Salary: $295K–$500K plus equity
Company
OpenAI is an AI research and deployment company focused on developing and safely deploying general-purpose artificial intelligence.
What you will do
Design and run experiments with frontier models and hands-on security techniques to discover and reproduce vulnerabilities in sandboxes, research infrastructure, and reachable services.
Develop threat models and adversarial tests for sandbox escapes, unauthorized agent communication, and access to sensitive data or systems through shared services.
Prototype and evaluate isolation defenses, layered protections, agent permissions, and tool configurations with infrastructure owners.
Red-team model-based safeguards and monitors, investigating evasion and measuring their ability to detect or prevent unsafe behavior.
Build automated adversarial tests and repeatable evaluations to verify fixes, assess generalization, and identify regressions as models and attack strategies change.
Partner with Research, Security, Agent Security, and infrastructure teams to prioritize risks and turn findings into deployed improvements.
Requirements
Hands-on research experience with large language models, including designing experiments and evaluating model behavior.
Ability to turn open-ended questions into hypotheses, build reproducible experiments at scale, and use results to develop or improve methods.
Ability to design rigorous evaluations, investigate unexpected results, and test findings across models, environments, and attack strategies.
Ability to write and debug code independently and build research tools and prototypes through experimentation.
Practical security experience in areas such as sandboxing, vulnerability research, adversarial testing, detection, or monitoring.
Ability to work with Security and infrastructure partners on threat models, findings, and deployed safeguards.
Nice to have
Experience with operating systems, containers or virtualization, distributed infrastructure, or AI agents.
Culture & Benefits
Work closely across Research, Security, and infrastructure organizations.
Build protections intended to keep increasingly capable models contained, observable, and under human control.
Full-time employment in a hybrid San Francisco role.
Equity is included in the compensation offer.
OpenAI is an equal opportunity employer and provides reasonable accommodations for applicants with disabilities.
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 →