Senior Data Scientist, Medical Imaging

Description

The company is a medical technology business focused on diagnosing and managing coronary artery disease, the number 1 cause of death worldwide.

Its flagship product is an AI-driven, non-invasive cardiac test that follows the ACC/AHA Chest Pain Guidelines. It builds a colour-coded 3D model of a patient's coronary arteries that shows how blockages affect blood flow to the heart.

The company offers an integrated, AI-driven, non-invasive heart care solution across the CCTA pathway. It helps clinicians find stenoses, measure coronary blood flow, and characterise and quantify atherosclerosis.

The company is publicly traded, supported by medical societies worldwide, and cleared for use in the US, UK, Europe, Japan and Canada. It has been used for more than 750,000 patients worldwide.

The product pipeline and the team are both growing.

The role is for a Senior Data Scientist with strong statistics and deep learning skills to lead the company's data-centric AI work. You will study how real-world imaging conditions and technical variables change image appearance, and how that in turn changes the accuracy and reproducibility of deep learning algorithms.

You will own the analysis pipeline that measures these effects, and you will test data-driven ways to correct them. You will work across teams to turn findings into algorithm and product decisions, including annotation protocols that produce high-quality labels for training and validation.

This suits someone who enjoys drawing insight from large, untidy real-world imaging data and turning it into clear, workable solutions.

Responsibilities

• Build reproducible analysis pipelines and visualisations over population-scale data to describe dataset distributions and show where models are weak across different patient groups.
• Explore, build and validate methods that harmonise complex data factors and image variation.
• Turn findings on data variance and model behaviour into data curation requirements, training robustness strategies and architecture recommendations for downstream algorithms.
• Write annotation protocols for algorithm training and validation, used for internal product work and FDA submissions.
• Work with research scientists, machine learning engineers, systems engineers, process engineers, product and regulatory teams. Turn messy data into clear analyses that guide decisions, and produce outputs other functions can use.

Requirements

• Hold a Master's or PhD in data science, computer science, medical image analysis, statistics, biomedical engineering or a related quantitative field.
• Have 5+ years of industry experience in data science, machine learning or image analysis, or 3+ years if you hold a PhD.
• Be fluent in reproducibility and agreement statistics, including variance components, Gage R&R, intraclass correlation, repeatability and reproducibility coefficients and Bland-Altman analysis. You can tell correctable bias apart from irreducible variance.
• Understand medical image data structures and the physical and clinical realities of imaging. Have experience with image processing tools and building algorithms on medical imaging data.
• Be expert in Python and statistical libraries such as pandas, scipy, statsmodels and seaborn or matplotlib. Have a track record with large, complex, messy datasets.
• Have hands-on experience developing or fine-tuning deep learning models, preferably in PyTorch, for segmentation, classification or detection on medical images.
• Use modern agentic tools and LLMs such as GitHub Copilot, Gemini or Claude every day to speed up software development, prototype data solutions and build reproducible pipelines.
• Explain complex, multi-dimensional analyses clearly and strategically to non-technical stakeholders.
• Preferred: published research on domain generalisation, image harmonisation or out-of-distribution detection in medical imaging.
• Preferred: experience with large multi-vendor CT imaging datasets, including acquisition and reconstruction protocols.
• Preferred: a proven record of diagnosing data and annotation problems behind algorithm performance gaps and designing data-driven fixes.
• Preferred: experience with large-scale data querying and cloud storage, such as AWS or SQL.
• Preferred: experience developing SaMD products and contributing to regulatory filings.
• Preferred: experience with biostatistics supporting FDA submissions.

Conditions

Base salary is estimated at $170,000 to $240,000.

The role is hybrid.

The company is an equal opportunity employer and does not discriminate on any basis protected by federal, state or local law.

Benefits

• Bonus and equity are offered on top of base salary.