12 дней назад
Large Language Model Intern (AI)
Вакансия из Hirify Global, списка международных tech-компанийДля мэтча и отклика нужен Plus
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Описание вакансии
Текст:
TL;DR
Large Language Model Intern (AI): Training and optimizing language models for Razer Synapse across corpus construction, fine-tuning, evaluation, and deployment optimization with an accent on production-scale experimentation and on-device constraints. Focus on building training datasets, running end-to-end model experiments, benchmarking quality and resource usage, and applying compression and quantization for device deployment.
Location: Singapore
Company
Develops gamer-focused software and hardware products, including the Razer Synapse device assistant.
What you will do
- Construct and prepare training corpora through cleaning, filtering, deduplication, and data mixing.
- Run supervised fine-tuning, LoRA, and preference optimization experiments.
- Evaluate model quality, latency, and memory usage against production constraints.
- Conduct compression and quantization experiments for on-device deployment.
- Own workstreams end to end, from experiment design and execution through debugging, analysis, and iteration.
- Collaborate with senior data scientists and engineers on the model powering the Razer Synapse device assistant.
Requirements
- Currently pursuing a Bachelor's, Master's, or PhD in Computer Science, AI, Data Science, or a related field.
- Knowledge of LLM training paradigms, including pretraining, SFT, LoRA, and preference optimization.
- Understanding of Transformer and deep learning fundamentals and the impact of latency and memory constraints on training decisions.
- Hands-on Python and PyTorch model training experience, including training-corpus data engineering and end-to-end experiment execution.
- Experience benchmarking model quality, latency, and memory; hands-on model training or fine-tuning is required.
- API-calling or prompt-engineering-only experience does not qualify.
Nice to have
- Model compression, quantization, or distillation experience.
- Distributed training, multi-GPU training, or parallelism experience.
- Experience with Hugging Face, Accelerate, DeepSpeed, or Linux.
- Training a model from scratch, on-device deployment work, or publications at venues such as CVPR, NeurIPS, ICML, ACL, ICLR, or EMNLP.
Culture & Benefits
- Work with a global team across five continents.
- Receive mentorship from senior engineers.
- Gain exposure to production-scale AI development and roadmap decisions for a shipping feature.
- Work in an inclusive and respectful environment 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 →