Роль четко сфокусирована на промышленной безопасности мультимодального ИИ в быстрорастущей исследовательской лаборатории, с высоким уровнем технической ответственности; обязательное присутствие в офисе важно учитывать при выборе.
Офис обязателен
Современный ИИ
Безопасность ИИ
Четкие задачи
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
TL;DR
Machine Learning Engineer (AI Moderation): Developing multilayered ML systems to detect harmful and illegal content across images, video, audio, and other modalities with an accent on VLM deployment, classification quality, and production scalability. Focus on integrating detection models into large-scale serving pipelines, tuning precision and recall, responding to moderation feedback, and designing technically effective safety policies.
Location: San Francisco, United States; hybrid office attendance required at least 2 days per week or one full week every other week.
Base annual salary: $220,000–$280,000 USD plus equity.
Company
A fast-growing research lab building foundational multimodal models for visual intelligence and generative media.
What you will do
Develop ML models and systems that detect harmful and illegal content across multiple classes and modalities.
Integrate moderation models into large-scale production serving pipelines with the infrastructure team.
Handle customer requests and feedback, making rapid adjustments to moderation behavior.
Collaborate with legal, policy, and product teams to create technically feasible moderation policies that preserve a positive user experience.
Requirements
Experience deploying VLMs at scale for multimodal content classification.
Experience building AI systems around state-of-the-art open-source VLMs.
Experience tuning models for classification, including building evaluation sets, reasoning about precision and recall thresholds, and monitoring model drift.
Comfort with the full ML engineering stack, from research and experimentation through large-scale deployment.
Ability to work as a Senior/Staff individual contributor across ML engineering and collaborate cross-functionally.
Ability to work from the San Francisco office at least 2 days per week or one full week every other week.
Nice to have
Experience with adversarial generative AI moderation, including prompt obfuscation, jailbreaks, output laundering, and evasion.
Moderation or safety experience across text, image, video, or audio.
Experience optimizing VLM inference latency at scale.
Experience with non-VLM safety classification solutions such as classical ML.
Knowledge of content copyright and/or child safety moderation systems.
Culture & Benefits
Work with a frontier research lab focused on deep technical understanding and high-quality products.
Values include being obsessed with the work, low ego, bold, and kind.
Full-time employment with base annual salary of $220,000–$280,000 USD plus equity.
AI-assisted tools may be used to organize applications, while hiring decisions are made by the team.
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 →