Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Описание вакансии
Текст:
TL;DR
Generative AI Intern (Large Language Models/PyTorch): Developing and optimizing large language models for intelligent and agentic AI applications with an accent on dataset preparation, fine-tuning, evaluation, and distributed training. Focus on benchmarking model quality and performance, applying quantization and distillation, and analyzing experiments for production-oriented Generative AI workflows.
Location: Singapore Razer AI Center
Company
Razer is a global gaming technology company developing products and software for gamers worldwide.
What you will do
- Prepare and validate datasets for language model fine-tuning and evaluation.
- Fine-tune large language models using LoRA, supervised fine-tuning, and preference optimization.
- Support distributed training and experimentation with PyTorch, Hugging Face, Ray, or DeepSpeed.
- Benchmark model quality, latency, memory usage, and training performance.
- Explore quantization and model distillation techniques, analyze results, and document findings.
Requirements
- Pursuing a Bachelor’s or Master’s degree in Computer Science, AI, Data Science, or a related field.
- Proficiency in Python and familiarity with PyTorch or TensorFlow.
- Knowledge of machine learning, deep learning, and Transformer models.
- Analytical, detail-oriented, and motivated to learn.
- Available for a minimum 16–20 week internship.
Nice to have
- Familiarity with Hugging Face, distributed computing, or Linux.
Culture & Benefits
- Opportunity to contribute to a global gaming mission and work with a team spanning five continents.
- Hands-on experience with large language model fine-tuning, evaluation, GPU computing, and distributed training.
- Exposure to dataset preparation, model optimization, experiment tracking, and production-oriented Generative AI workflows.
- Inclusive and respectful workplace with reasonable accommodations where needed.
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 →