Research Intern (AI)
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Описание вакансии
TL;DR
Research Intern (AI): Contributing to the pre-training and mid-training of large multimodal foundation models with an accent on high-throughput data pipelines and MLLM inference optimization. Focus on developing and deploying frontier generative AI technologies in real-world systems alongside top-tier research talent.
Location: Must be based in or able to work on-site at the Palo Alto, CA headquarters
Company
is a creative infrastructure company building real-time, multimodal generation and intelligent agentic platforms to democratize generative technology.
What you will do
- Contribute directly to large-scale pre-training and mid-training efforts for multimodal foundation models.
- Design, prototype, and optimize algorithms for high-throughput data pipelines and MLLM inference.
- Work with diverse data types including text, image, audio, and video to support generative model training.
- Collaborate with experienced researchers and engineers on state-of-the-art core technologies.
- Apply frontier research techniques to production-scale AI workflows.
Requirements
- Currently enrolled in a BS, MS, or PhD program with a focus on machine learning, deep learning, or multimodal AI.
- Proficiency in Python and ML frameworks such as PyTorch or TensorFlow.
- Ability to adapt quickly to fast-paced research environments and new technologies.
- Strong communication and cross-functional collaboration skills.
- Must be able to work on-site in Palo Alto, CA.
Nice to have
- Prior research or project experience in MLLM inference or generative models.
Culture & Benefits
- Flexible internship terms with both part-time and full-time options.
- Mentorship from world-class researchers and engineers.
- Competitive compensation.
- Dynamic, mission-driven startup environment.
- Opportunity to make an immediate impact on frontier AI 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 →