aillmmachine learningresearchreinforcement learningai safetyexperimental designhuman ai interactionai evaluation
Вакансия из Hirify Global, списка международных tech-компаний Для мэтча и отклика нужен Plus
Пожаловаться
Зарплата и рынок
Вышеразвернуть
$275к-400к/год
$200к$250кмедиана$270к
По похожим ролям (49) ~ $250к/год
91
Крутая вакансия
развернуть
Чётко очерченная руководящая роль в исследованиях с доступом к передовым AI-проектам, ресурсам для исследований и возможностью влиять на реальные продукты.
Передовой AI
Ресурсы для исследований
Понятная зона ответственности
Выше рынка
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
TL;DR
Director of AI Research (AI evaluation and human data): Leading research on benchmarks, evaluations, environments, datasets, and human-AI interaction for frontier AI with an accent on experimental design, measurement, safety, and expert data. Focus on setting research direction, building a senior UK/US team, partnering with frontier labs and regulators, and turning scientific methods into tools and evidence customers use.
Location: Remote, based within commutable distance of at least one global hub in London, New York, or San Francisco.
Base salary: $275,000–$400,000 annually for the full-time role in San Francisco or New York.
Company
hirify.global provides a platform connecting researchers and companies with a global participant pool for high-quality, ethically sourced human behavioral data and feedback used in AI development.
What you will do
Set research strategy for more capable and safer AI systems, trustworthy evidence, human signal, and public-benefit research.
Lead work on expert data, agentic training and evaluation environments, AI safety, human-AI interaction, model capability measurement, and task and data design.
Build reusable benchmarks, evaluations, environments, datasets, methods, tools, and products adopted by frontier labs and customers.
Partner with product, engineering, commercial, and research leaders to take methods to market and report customer, revenue, and scientific impact.
Represent hirify.global with frontier labs, regulators, safety institutes, academia, and the research community.
Hire, lead, and grow a UK/US team of research engineers, research scientists, and forward-deployed researchers.
Requirements
Research credibility in post-training data, reinforcement learning environments, AI evaluation, or AI safety and human-AI interaction, demonstrated through papers, benchmarks, or methods recognized or used by frontier lab researchers.
Strong grounding in experimental design and measurement with human participants, including sampling, validity, reliability, and bias.
8+ years of research experience, including 3+ years leading researchers across locations through hiring, management, and development of senior staff.
Track record of turning research into customer-used results and adopted artifacts such as benchmarks, datasets, environments, or tools, not only papers.
Clear perspective on the future of data and evaluation for frontier AI, with judgment to make portfolio decisions and communicate them to executives, customers, and the research community.
Hands-on ability to review methods, read code, and challenge results.
Nice to have
Experience at a frontier lab, AI safety institute, human data company, or evaluation company.
Background spanning machine learning and behavioral or social science, including psychology, psychometrics, economics, or computational social science.
Experience building agentic evaluations or LLM-as-judge pipelines at scale.
Experience working with regulators or standards bodies on AI evaluation.
Culture & Benefits
Remote working within a mission-driven AI research organization, subject to proximity to a global hub.
Access to a human data platform, compute, budget, and tooling for large human studies and model experiments.
Competitive salary and benefits.
Protected time for novel and public-benefit research.
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 →