Machine Learning Intern (AI)
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Описание вакансии
TL;DR
Machine Learning Intern (AI): Designing, training, and improving frontier AI models for content generation, semantic search, and RAG with an accent on large-scale distributed training and novel research implementation. Focus on building elegant training/deployment pipelines and developing techniques to serve models safer, better, and faster.
Location: Remote (Must be based in Canada, Europe, Korea, United Kingdom, or United States)
Company
trains and deploys frontier models for developers and enterprises to power AI systems like semantic search, RAG, and agents.
What you will do
- Design, train, and optimize cutting-edge machine learning models.
- Develop new techniques to train and serve models safer, better, and faster.
- Execute large-scale training on massive datasets.
- Explore active and continual learning strategies for streaming data.
- Collaborate with product teams to develop AI-driven solutions.
Requirements
- Currently enrolled student in a post-secondary program.
- Available for a full-time 3-6 month internship, co-op, or research term.
- Proficiency in Python and ML frameworks such as Tensorflow, JAX, and XLA/MLIR.
- Experience with large-scale distributed training strategies and Transformers.
- Strong communication and problem-solving skills.
- Demonstrated passion for applied NLP models and products.
Nice to have
- Experience writing GPU kernels using CUDA.
- Experience training models on TPUs.
- Publications in top-tier venues like NeurIPS, ICML, ICLR, or ACL.
Culture & Benefits
- Open and inclusive culture working with a world-class AI research team.
- Full health and dental benefits, including a dedicated mental health budget.
- Remote-flexible arrangement with a co-working stipend.
- Generous vacation policy providing 6 weeks of leave per year.
- 100% Parental Leave top-up for up to 6 months.
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 →