About the role

Join Tether's AI model team to research and engineer large language and multimodal model architectures. The work focuses on large-scale pre-training, multimodal data alignment, model efficiency, and reliable distributed training.

Responsibilities

• Run foundational pre-training for LLM and multimodal models across large multi-node NVIDIA GPU clusters.
• Design and scale model architectures, tokenizers, and cross-modal alignment layers.
• Source, filter, and curate large text and multimodal datasets and build efficient data pipelines.
• Plan experiments, analyze results, improve token efficiency, and resolve training and alignment bottlenecks.
• Improve distributed training scalability and hardware efficiency.

Requirements

• Degree in computer science or a related field; a PhD in NLP or machine learning is preferred.
• Hands-on experience with large-scale LLM or multimodal pre-training and distributed training frameworks.
• Strong knowledge of transformer and non-transformer architectures.
• Strong PyTorch and Hugging Face experience across model development, continual pre-training, and deployment.

How to apply

Apply through the official Tether careers page linked to this opportunity.