Hiring nowMedical imaging and diagnostic devices

Digital Pathology AI Scientist

Also called Computational Pathology Scientist

A Digital Pathology AI Scientist builds models that read tissue images and proves they hold up in the real world. That matters because a diagnostic product only sells if pathologists, regulators and laboratories trust it.

What they do

They build and test models on digitised tissue slides, working from data preparation through to a reviewed study. They agree labels and ground truth with pathologists, and they check what the model has actually learned, not just its headline score. They study how stain, scanner, slide preparation and site differences change results, and they design ways to cope with that shift. They measure uncertainty, look at failure cases with clinicians, and write up methods so quality and regulatory colleagues can follow them. They also keep records that stand up to audit.

What a strong hire delivers

In the first months, a strong hire delivers a working pathology model with a clear account of what it does and does not do. They set up a validation plan that covers several laboratories, scanners and patient subgroups. They produce a study that a pathologist has reviewed and a regulatory colleague can use. The team also gets a clearer view of the main risks to performance outside the development data.

How you can tell they are good

Ask them to walk through a model they built and where it failed. A strong candidate talks about slide-level problems, label disagreement between pathologists and site effects before they mention architecture. They can explain how they tested on unseen laboratories and scanners. They are specific about uncertainty and when the model should defer to a human. Be wary of anyone who quotes a single accuracy figure and treats validation as a final step.

Why it is hard to fill

The work sits between machine learning, pathology and regulated device development, and few people have all three. Titles vary widely, so searching on one phrase misses most of the field. Many strong ML people have never dealt with stain variation or clinical review. Many pathology-literate scientists have limited experience of rigorous model validation. Candidates who have worked with real multi-site data, and seen a study reviewed, are rare and are often settled.

Where to find them

Look in computational pathology groups in academic medical centres, and in diagnostics and medical device companies building image-based products. Others come from biomedical engineering, medical imaging research and clinical science, or from machine learning teams that moved into healthcare. Related titles include computational pathology scientist, medical imaging scientist and clinical ML researcher. Companies with regulated imaging products are a strong source because their people already work within quality systems.

Indicative pay

Not enough published evidence yet to give a fair range for this role, which is common for work this new. We benchmark it for you as part of a search.

Interview questions for a Digital Pathology AI Scientist

  1. Tell us about a pathology or medical imaging model you took from data to validation. What went wrong along the way?

    A strong answer gives concrete problems such as label noise, site bias or poor slide quality, and says what they changed. It shows ownership of the whole path. Worry if they only describe architecture or report a single score.

  2. How would you test whether a model trained at one laboratory will work at another?

    They should cover external held-out sites, scanner and stain differences, and subgroup analysis. Good answers include stain handling and checks on what the model relies on. Worry if they suggest a random split of the same data.

4 more questions, with what a strong answer covers, are in the Role Kit.

Job description you can post

We are building AI-enabled diagnostic functions that read tissue images. We need a scientist who can develop pathology models and show, with pathologist-reviewed evidence, that they perform across laboratories, scanners and patient groups. You will work with clinicians, engineers and quality colleagues.

What you will do

  • Develop and train models on digitised histopathology images
  • Design validation studies across sites, scanners and patient subgroups
  • Work with pathologists on labels, ground truth and review of results
  • Investigate and reduce the effects of domain shift
  • Quantify model uncertainty and define when cases need human review
  • Document methods and results to support quality and regulatory needs

What you will bring

  • Demonstrated work building computer vision models on medical or tissue images
  • Experience validating models on external or multi-site data
  • Clear understanding of stain, scanner and slide variation
  • Experience working with clinicians to define ground truth
  • Ability to explain results and limits to non-specialists
  • Comfort working within quality or documentation controls

Nice to have

  • Experience with regulated medical device development
  • Published or reviewed clinical validation work
  • Familiarity with uncertainty quantification methods

Free to use and adapt. If it saves you time, a link back to this page is appreciated.

Skills that matter

  • Computer visionAdvancedAnalyse or generate visual information
  • Clinical validationAdvancedAssess intended clinical benefits, harms and performance
  • AI governance and assuranceWorkingAssign accountability and verify lifecycle controls
  • Uncertainty quantificationWorkingEstimate confidence, calibration and model uncertainty

Other roles in medical imaging and diagnostic devices

Hiring nowMedical Imaging AI EngineerA Medical Imaging AI Engineer builds image-analysis functions for a defined clinical use and shows they work on representative clinical data. They…Hiring nowAI Medical Device Validation EngineerAn AI Medical Device Validation Engineer tests device software and judges whether the evidence supports its intended use. They help a company show…EmergingAI Device Post-Market SpecialistAn AI Device Post-Market Specialist watches how an AI-enabled diagnostic or monitoring device performs once it is in clinical use. They catch drift…

Related roles in other families

Hiring nowComputer Vision EngineerComputer vision and spatial perceptionA Computer Vision Engineer builds detection and recognition systems that work in a defined setting, and proves how well they work. Companies rely on…Hiring now3D Vision ResearcherComputer vision and spatial perceptionA 3D Vision Researcher recovers spatial structure from images and depth data, then proves it holds up. Companies building robotics, mapping, AR or…Hiring nowVideo Understanding ScientistComputer vision and spatial perceptionA Video Understanding Scientist builds models that recognise actions, events and the dependencies between them in video. Companies hire for this…EmergingOpen-World Perception ResearcherComputer vision and spatial perceptionAn Open-World Perception Researcher builds vision models that know when they are looking at something unfamiliar. That matters because a system that…

How we wrote this

Role definitions are our own editorial work, checked against occupational classifications, standards, research and employer pages. Titles vary by company, and this is a description of the work, not a job advert.

← Medical imaging and diagnostic devicesAll new job families

Start a search

Hiring a Digital Pathology AI Scientist?

Tell us the work. We find the people already doing it, whatever their job title says, and you meet them already interviewed for your role.