Centaur Labs

Labeling medical images at scale

Machine Learning Engineer at Centaur Labs

Location
Boston, MA / Remote
Job Type
Full-time
Experience
1+ years
Connect directly with founders of the best YC-funded startups.
Apply to role ›
Erik Duhaime
Erik Duhaime
Founder

About the role

You will play a key role in the success of a collaboration between Centaur Labs and Brigham and Women's Hospital (BWH) to develop large-scale labeled datasets of ultrasound clips, as well as state-of-the-art deep learning algorithms trained on those datasets.

You will be employed by Centaur Labs, but you will also have a postdoc appointment with and work closely with researchers at BWH.

Your objectives will include:

  • Creating labeling pipelines and procedures that will produce high-quality, clinically relevant labels for AI development.
  • Development and validation of state-of-the-art deep learning algorithms.
  • Serving as a project manager overseeing work at BWH to de-identify and clip ultrasound videos, and then secure transfer of clips to Centaur Labs for labeling.
  • Conducting IRR analyses on labelers from the Centaur Labs network to identify optimal methods for statistically weighting and combining multiple rater opinions.
  • Manuscript writing and preparation, including work on releasing deep learning algorithms in standardized formats to make them widely accessible to the broader machine learning and healthcare communities.

Requirements

  • PhD/MD. This is a postdoc position, and applicants must have either an MD or a PhD in data science, computer science, or a degree that involved large scale medical data analysis (e.g., neuroscience, computational biology, etc.).
  • Required skills include SQL, Amazon Web Services (AWS), and deep learning experience with medical image and/or video datasets.

Preferred Skills

  • Experience building computer vision models.

Why you should join Centaur Labs

At Centaur Labs, we label data at scale to enable breakthroughs in medical AI. Our novel approach uses collective intelligence to aggregate medical opinions from experts and students all over the world. Our labeling customers include leading AI startups and prominent research organizations. We are a small, tight-knit team headquartered in Boston, and we're funded by Y Combinator, Accel, and GFC.