Research Scientist, Multimodal Foundation Models

Elorian; Iowa; In loco; USD 140.000 - 190.000; Tempo pieno

Descrizione del lavoro

Elorian is seeking a PhD-level researcher to lead frontier AI research across foundation models, multimodal reasoning, and distributed training. You will formulate hypotheses, run large-scale experiments, and translate breakthroughs into scalable capabilities for real-world use.

Collaborations across research and systems engineering teams will turn novel architectures and data-centric methods into robust models deployed at scale, with dedicated compute and engineering support.

Competenze

• Ph.D. or equivalent industry research track record in ML, CV, NLP, or related quantitative discipline.
• Deep understanding of deep learning, transformers, and large-scale training dynamics.
• Proficiency in Python and modern DL frameworks (PyTorch or JAX).
• Ability to define research agendas, design rigorous experiments, and solve frontier problems.

Mansioni

• Drive frontier research at the intersection of foundation models, multimodal understanding, and complex visual reasoning.
• Design and prototype novel model architectures and scalable training methodologies.
• Plan and execute large-scale distributed training experiments on accelerator clusters.
• Advance data-centric intelligence, synthetic data generation, and rigorous benchmark evaluation.
• Collaborate with cross-functional teams to translate research breakthroughs into scalable capabilities.

Conoscenze

Python programming

Deep learning fundamentals

Transformer models

Multimodal understanding

Empirical research

Formazione

Ph.D. in ML/CV/NLP or related field

Strumenti

PyTorch

JAX

Megatron-LM

DeepSpeed

Ray

Descrizione del lavoro

Elorian is seeking a PhD-level researcher to lead frontier AI research across foundation models, multimodal reasoning, and distributed training. You will formulate hypotheses, run large-scale experiments, and translate breakthroughs into scalable capabilities for real-world use.

Collaborations across research and systems engineering teams will turn novel architectures and data-centric methods into robust models deployed at scale, with dedicated compute and engineering support.