Staff Data Scientist, Machine Learning

$200,000 - $227,000/Yr

Valo Health - Lexington, MA

Full-time - Mid Level

Lexington, MA

Administrative and Support Services

About the position

As a Staff Scientist in Machine Learning at Valo Health, you will play a pivotal role in a dynamic team dedicated to revolutionizing the drug discovery and development process. Your primary focus will be on building and enhancing a powerful computational platform that leverages machine learning (ML) to address complex scientific challenges involving clinical and biomedical data. You will collaborate closely with a diverse group of professionals, including data scientists, biological scientists, epidemiologists, and software engineers, to create innovative solutions that push the boundaries of traditional industry practices. In this role, you will engage in hands-on exploratory analysis and modeling of high-dimensional longitudinal data, generating actionable insights that inform decision-making across multiple projects. You will design and implement cutting-edge ML approaches utilizing Valo's proprietary platform, as well as newly published methodologies, to tackle scientific problems effectively. Your contributions will extend to the development of robust, automated pipelines that incorporate deep learning and classical ML techniques, particularly in the context of high-dimensional electronic health records and omics data. Additionally, you will be responsible for creating well-designed, tested, and documented software packages that enhance the team's capabilities. Your role will also involve planning, executing, interpreting, and communicating results to ensure alignment with key stakeholders and the relevance of your models and analyses. As an active team member, you will participate in code, design, and analysis reviews, fostering a collaborative environment that encourages innovation and excellence.

Responsibilities

• Perform hands-on exploratory and defined analysis and modeling of high-dimensional longitudinal data to generate fit-for-purpose evidence for decision making for multiple projects.
• Design and implement innovative ML approaches leveraging Valo's proprietary platform as well as newly published methods.
• Contribute to the design and implementation of robust, automated pipelines leveraging deep learning and classical ML on high dimensional electronic health records and omics data to solve scientific problems.
• Develop well-designed, tested, and documented software packages.
• Contribute to planning, execution, interpretation, and communication of results.
• Collaborate with cross-functional teams and key stakeholders to derive user requirements, maintain alignment, and ensure the relevance and impact of models, analyses, and pipelines.
• Be an active team member in code, design, and analysis review.

Requirements

• BS, MS, or PhD in a quantitative field.
• Broad experience in ML including random forest, logistic regression, dimensionality reduction, clustering, metrics, model selection, feature selection, and explainability.
• Demonstrated experience implementing and applying deep learning approaches including representation learning on high dimensional or multimodal data.
• Proficient in Python and experience with ML, deep learning, and data science packages (e.g., scikit-learn, pytorch, statsmodels, scipy, MLlib).
• Experience with large-scale data analytics engines (e.g., Spark or Dask) and working in cloud environments (e.g., AWS).
• Experience with collaborative software development using source control management (e.g., git, unit testing, code review, CI/CD).
• Experience with MLops including workflow orchestration (e.g., Airflow, Prefect), experiment tracking (e.g., MLflow), containerization (e.g., Docker), and reproducible research.
• Experience with statistical methods such as hypothesis testing, longitudinal modeling, and time to event analysis.
• Ability to work effectively in teams or independently.
• Strong work ethic with a bias for execution and an ability to manage multiple priorities, ambiguity, and tight timelines.
• Experience with electronic health records and omics data preferred.
• Familiarity with the drug discovery and development process preferred.

Nice-to-haves

• Experience with electronic health records and omics data preferred.
• Familiarity with the drug discovery and development process preferred.