In Brief

• We're an early-stage startup on a mission to make healthcare proactive by empowering physicians, nurses, and care team members with real-time data to save lives.
• We’re looking for an experienced full-stack Data Scientist with expertise in state-of-the-art ML methods who wants to own the end-to-end effectiveness of their models in a live, clinical AI product.
  
  
  

Who We Are
Bayesian Health’s mission is to improve patient outcomes by empowering clinicians with the insights they need to make the right decision for the right patient at the point-of-care. We’re a diverse team of clinicians, engineers, machine learning experts, product designers, and performance improvement leaders committed to enabling smarter, patient-specific care delivery through unlocking the power of data.

What makes us different? We’re results oriented and outcomes obsessed. We also understand that to address our users’ needs, we must go beyond building great technology. After spending a decade working side by side with top clinicians from three of the world’s leading health systems, we know what it takes to develop usable, high-value, clinically-relevant solutions. We are experts at rummaging through messy, incomplete, biased datasets. We are able to make sense of, and continuously learn from data, to help our physicians practice more precise medicine.

Read more about our recent publication in Nature Medicine that associates our products with lives saved.

What You’ll Do
As a full-stack Data Scientist, you will own the effectiveness of models that you develop and maintain, throughout the life cycle of production ML systems.

You will:

• Develop and tune innovative, new ML models and labeler systems based on deep understanding of clinical use cases and state-of-the-art ML methods.
• Define and implement model/algorithm improvements to refine the behavior of existing ML-based systems.
• Identify strategies for improving our production ML-based systems, and write, debug, and deploy production-grade Python code to implement those strategies. 
• Produce high-quality, one-off, aggregate analyses of model behavior, including producing high-quality studies for publication.
• Design and create proofs of concept of innovative and original ML solutions to clinical challenges.
• Investigate to understand and communicate model behavior in specific cases.
• Write scripts to perform calculations for internal/external reporting and dashboards, to enable investigation of model performance (individual cases as well as statistics).
• Investigate data coming from EHRs and build high-quality data mapping code.
• Be an integral part of our diverse, inclusive, collaborative, fully-remote technology team.

Who You Are
As a full-stack Data Scientist, you are not satisfied with training and tuning ML models that predict clinical conditions in patients; you also want to own the effectiveness of your model in the real world. In practice, that means you aren’t afraid to get your hands dirty by writing data mapping code, debugging a specific patient case by following patient data as it moves through our AWS services, or improving the timeliness of your model’s predictions by reading and writing production-grade Python and SQL code.

• You either have a Ph.D. in a relevant field plus 3+ years relevant experience, or a relevant Master’s degree and 5+ years experience as a full-stack data scientist. Relevant degrees include Machine Learning, Computer Science, Statistics, and Biostatistics.
• You are an expert in state-of-the-art machine learning approaches, and more importantly, you stay up to date on relevant works in the literature to guide your design and development strategy.
• You have experience writing production-grade Python and SQL code to implement and evaluate ML models in production systems. 
• You have experience in the design and execution of health outcomes/effectiveness studies and in relevant statistical methods.
• You have experience using messy clinical data to solve health care challenges. In this experience you have built a solid understanding of health care, clinical data, and clinical coding standards such as SNOMED and ICD.
• You have experience using MLOps tools such as SageMaker and MLFlow.
• You are passionate about making clinical care more effective and efficient.
• You believe data has a larger role to play at the point-of-care, and want to make that a reality.
• You have a track record as a good communicator and a collaborative, driven team player.