Research Scientist / Research Engineer - Foundations, Search

What you'll do

• Design embedding models, retrieval systems, and indexing methods for grounding, relevance, and adaptive reasoning.
• Build and integrate infrastructure for training, evaluating, and deploying embeddings in frontier models.
• Advance dense, sparse, and hybrid representations, metric learning, and learning-to-retrieve systems.

What they ask for

• Proven experience leading high-performance teams in ML infrastructure or foundational research.
• Deep expertise in representation learning, embedding models, or vector retrieval systems.
• Familiarity with transformer-based LLMs and how embedding spaces interact with language-model objectives.
• Research experience in contrastive learning, embedding learning, or metric learning.

Nice to have

• Experience building or scaling large machine-learning systems, especially embedding pipelines.
• First-principles approach to retrieval and memory for large models.

Pay

$380K – $850K USD base salary

Tools

transformer-based LLMs

AI focus

Develop search, retrieval, and learned-representation capabilities that let models access relevant information and support memory and knowledge access.