Currently enrolled in a PhD program in machine learning, computer science, electrical engineering, computational neuroscience, physics, or a related field.
Available full time for six months.
Strong Python and hands-on PyTorch development, or an equivalent framework such as JAX or TensorFlow.
Experience taking a model from a paper or prototype to something that trains reliably on real data.
Comfort with large numerical datasets and memory-mapped arrays.
Solid grounding in first principles, such as linear algebra and signal processing.
Good software engineering practices, including version control, testing, and code review.
Self-directed and able to execute on a goal rather than needing a step-by-step guide.
Responsibilities
Design, implement, train, and evaluate models for one part of the pipeline: functional ultrasound reconstruction, signal preprocessing, or downstream decoding.
Run experiments by forming and testing hypotheses about why a model is not learning, designing ablations, and tuning architectures, losses, and training schedules.
Extend models to new data regimes, such as 2D to 3D volumetric data, new probes, or new subjects, and build the preprocessing needed to train on them.
Contribute to training and evaluation infrastructure, including experiment tracking, metrics, and regression checks that assess whether a model is better.
Debug numerical and performance problems, such as NaNs under mixed precision, silent shape bugs, and GPU memory ceilings.
Work directly with imaging scientists and neuroscientists to understand data sources and artifacts.
Reproduce results from relevant literature and present findings to the team.
Desired Qualifications
Experience with neural operators (FNO), implicit neural representations, or physics-informed machine learning architectures.
Exposure to neural decoding, brain-computer interfaces, or neuroimaging data.
Experience with mixed-precision or multi-GPU training.
Publications at machine learning or neuroscience venues, or contributions to open-source machine learning tooling.
About Bridge Neurotech
Bridge Neurotech is building a non-invasive functional ultrasound imaging (fUSI) platform with AI-based neural decoding for brain-computer interface applications. We are a seed-stage neurotechnology company developing technology that has the potential to transform how we interact with the human brain.
About the role
We are looking for a PhD student in machine learning or a related field to join Bridge for a six-month internship or co-op, working on the models and infrastructure behind our neural decoding pipeline: from raw functional ultrasound acquisition through image reconstruction, preprocessing, and decoding.
Recent work has shown that brain activity can be mapped into the latent spaces of pretrained deep learning models to reconstruct visual perception and continuous language. This has been demonstrated with fMRI, in research settings. Our work is to bring these techniques to functional ultrasound: a modality with superior spatial and temporal resolution, portability, and a path to consumer deployment.
You will own a well-scoped project from open question to result, working closely with our ML/AI Engineering Lead.
What you’ll do
Design, implement, train, and evaluate models for one part of the pipeline: fUSI reconstruction, signal preprocessing, or downstream decoding.
Run the experiments: form and test hypotheses about why a model is not learning, design ablations, and tune architectures, losses, and training schedules.
Extend models to new data regimes, for example 2D to 3D volumetric data, new probes, or new subjects, and build the preprocessing needed to train on them.
Contribute to our training and evaluation infrastructure: experiment tracking, metrics, and the regression checks that tell us when a model is actually better.
Debug numerical and performance problems, for example NaNs under mixed precision, silent shape bugs, and GPU memory ceilings.
Work directly with our imaging scientists and neuroscientists to understand where the data comes from and what its artifacts mean.
Reproduce results from the literature that matter to our roadmap, and present what you find to the team.
Who you are
Required
Currently enrolled in a PhD program in machine learning, computer science, electrical engineering, computational neuroscience, physics, or a related field.
Available full time for six months.
Strong Python and hands-on PyTorch development (or an equivalent framework such as JAX or TensorFlow).
Experience taking a model from a paper or a prototype to something that trains reliably on real data.
Comfort with large numerical datasets and memory-mapped arrays.
Solid grounding in first principles, for example linear algebra and signal processing.
Good software engineering practices: version control, testing, and code review.
Self-directed and self-motivated, with the ability to execute on a goal rather than a step-by-step guide.
Nice to have
Experience with neural operators (FNO), implicit neural representations, or physics-informed machine learning architectures.
Exposure to neural decoding, brain-computer interfaces, or neuroimaging data.
Experience with mixed-precision or multi-GPU training.
Publications at machine learning or neuroscience venues, or contributions to open-source machine learning tooling.
What we offer
A six-month project with real ownership, scoped with you around our decoding roadmap.
Mentorship from our ML/AI Engineering Lead, and direct collaboration with leading clinicians and neuroscientists at top-tier academic institutions.
Work on a modality early enough that the tooling does not exist yet. Much of what you build will be the first of its kind.
Bridge Neurotech is an equal opportunity employer. We encourage candidates of all backgrounds to apply.