AI Research Engineer (Pre-training - LLM & Multi-Modal)

Tether Operations Limited is at the forefront of the financial revolution, pioneering global blockchain solutions and cutting-edge AI research. We are seeking an AI Research Engineer to join our model team and drive innovation in architecture development for large-scale and multi-modal systems.

Key responsibilities

• Conduct foundational pre-training for LLMs and multi-modal models on large-scale distributed clusters with thousands of NVIDIA GPUs.
• Design and prototype innovative architectures, tokenizers, and cross-modal alignment layers to enhance model intelligence.
• Source, filter, and curate massive-scale datasets while establishing robust pipelines for efficient pre-training.
• Execute experiments and refine training methodologies to optimize performance and token efficiency.
• Investigate and resolve bottlenecks in model efficiency, computational performance, and training stability.

Requirements

• Degree in Computer Science or a related field, ideally a PhD in NLP or Machine Learning with a strong track record in AI R&D.
• Hands-on experience with large-scale LLM or multi-modal pre-training runs on distributed systems.
• Deep knowledge of state-of-the-art transformer architectures and modifications for scalability.
• Strong expertise in PyTorch and Hugging Face libraries with practical experience in model development.
• Familiarity with large-scale distributed training frameworks and tools.

What we offer

• The opportunity to work within a global talent powerhouse on groundbreaking AI research.
• A fully remote work environment that fosters collaboration with some of the brightest minds in the industry.
• The chance to shape the future of digital finance and AI technology at a leading global organization.