Senior ML Engineer - Foundation Models for Brain Data

Protocol Labs, Inc.; San Francisco (CA); On-site; USD 180,000 - 240,000; Full time

Job summary

Protocol Labs, Inc. is hiring a Senior Machine Learning Engineer to design and train foundational models across neural data and behavior. You will work on the core generative model, handle end-to-end training, and shape modeling direction with a small team in San Francisco.

Focus areas include architecture, scalable training, and representation learning across EEG/MEG/fMRI signals. Strong collaboration with data and research teams is expected to deliver production-ready, well-documented code.

Qualifications

• Trained large deep learning models end-to-end.
• Experience training transformer or large sequence models with distributed training.
• Proficiency in Python and PyTorch, with reproducible model code.
• Strong prototyping across research and engineering at scale.
• Comfort working with multimodal or time-series data.
• Ability to learn new domains quickly from literature.

Responsibilities

• Design, train, and iterate on large generative models over multimodal neural data.
• Own training at scale: data loading, distributed multi-node training, hyperparameter optimization.
• Develop representations that capture structure across species and modalities.
• Train models on neural and behavioral data, including EEG/MEG/fMRI related signals.
• Deliver modular components of the modeling stack as working, documented code.

Skills

Deep learning

Transformer

Distributed training

Python

PyTorch

Multimodal data

Time-series

Research & engineering

End-to-end ownership

Experiment design

Communication

Job description

Protocol Labs, Inc. is hiring a Senior Machine Learning Engineer to design and train foundational models across neural data and behavior. You will work on the core generative model, handle end-to-end training, and shape modeling direction with a small team in San Francisco.

Focus areas include architecture, scalable training, and representation learning across EEG/MEG/fMRI signals. Strong collaboration with data and research teams is expected to deliver production-ready, well-documented code.