Senior/Principal Machine Learning Engineer

Boston or Remote ·  Full time

Beacon Biosignals is seeking a Senior/Principal Machine Learning Engineer to join us in our mission to make brain monitoring easily accessible, interpretable, and actionable.  You’ll work directly with Beacon Labs project teams composed of research scientists, neuroscientists and data scientists to deliver high-quality, reproducible analysis products that drive advancements in brain health, and

• …discover and formulate techniques for analysis that haven’t even been attempted with EEG outside of Beacon due to lack of data and/or infrastructure (e.g. “How might we refactor one of our deep learning-based feature detection algorithm to achieve high inter-rater agreement on noisy signals across a reduced set of channels?”, “What data augmentation techniques or validation strategies might we try in order to assess or reduce bias in real-world environments/specific patient populations”, etc.).
• …steer engineering efforts to standardize, improve, and automate our model development/deployment lifecycle and related tooling.
• …write modular and composable code with well-defined interfaces.
• …have the opportunity to co-author scientific papers whose impact pushes the clinical, neuroscience, and machine learning communities past contemporary limitations.

About you:

• You have 5+ years of relevant industrial data science experience and are well-versed in at least one DL/ML framework (e.g. Tensorflow, PyTorch, Flux).
• You have extensive experience working with time series data, preferably with healthcare biosignal data (e.g. EKG, actigraphy, EEG, Polysomnography, etc.).
• You’re comfortable building tools and overseeing experiments that uniquely leverage data from our clinical partners, public datasets, and our own sensors to aid clinical decisions.
• You realize that lowering hypothesis test turnaround time from months to minutes requires applying best-of-breed DevOps concepts to the machine learning workflow.
• You’ve leveraged cloud infrastructure to train models on inconveniently large datasets.
• You’re painfully aware that thoughtful data curation is a prerequisite to analysis results that hold up under practical application.
• You make use of best-of-breed experiment tracking systems to ensure that model performance is reproducible, observable, and debuggable.
• You can’t wait to combine your background with our unprecedented EEG dataset to blow published statistical EEG results out of the water.
• You’ve developed a solid, heavily-exercised workflow for debugging model performance; you now know more about the limitations of automatic differentiation and distributed heterogeneous computing than you ever thought you would when you started your ML journey (e.g. “How can I alter this model’s architecture to mitigate the decrease in sample throughput once I lower the minibatch size?”).
• You can ensure that we properly select and rigorously measure the quantities we’re optimizing for, both in our research and in our business.
• You enjoy working in teams and providing/receiving regular feedback to improve technical and operational practices.
• You have excellent speaking, listening and writing skills.
• You’re excited to work in an environment that makes heavy use of Julia, Docker/K8s, workflow orchestration tools (e.g. Argo, Airflow), SQL, and GraphQL.

Not sure if you meet 100% of our qualifications? There are numerous reports that systemically marginalized people (e.g. women) are less likely to apply for a job, but are more likely to be hired if they do apply. If you think you could thrive in this role, we encourage you to apply even if not every bullet perfectly describes you.

Beacon Biosignals is an equal opportunity employer; we do not discriminate on the basis of race, religion, color, national origin, gender, sexual orientation, age, marital status, veteran status, or disability status.

Tagged as: EEG, Machine Learning