Data Annotation Analyst

Ironsides Medical is seeking an annotation analyst to label clinical video and image data used to train and evaluate our machine learning models. The work is done in V7 Darwin, a collaborative annotation platform, and follows a written annotation protocol maintained with our engineering and clinical teams. This is a hands-on role where consistency and judgment on difficult cases matter more than speed alone. Note that onboarding and team meetings will be in person in Somerville, MA. Some work can be done remotely. Please only apply if you can be in person in Somerville, MA, as needed.

RESPONSIBILITIES Annotate video frames and image sets in V7 Darwin according to our annotation protocol Maintain consistency across large batches and across successive versions of the label definitions Flag ambiguous, low-quality, or out-of-scope cases for review rather than guessing at them Run quality-control passes on completed annotations, correct errors, and re-review as needed Track progress against weekly targets and raise blockers early Help refine the annotation protocol and edge-case guidance as the dataset grows Organize and export labeled datasets, keeping versioning and file structure clean and reproducible Handle clinical data in line with our confidentiality and data-handling requirements

REQUIRED QUALIFICATIONS Consistent, careful attention to detail over long sessions of repetitive visual work Ability to follow a written protocol precisely, and to ask rather than assume when a case is unclear Comfort learning an annotation platform quickly — V7 Darwin, or equivalent experience with CVAT, Labelbox, or SuperAnnotate Basic Python or scripting for batch operations, exports, and simple data checks Clear written communication for documenting decisions and edge cases

Prior image or video annotation work for machine learningBackground in biomedical engineering, life sciences, or a clinical field Familiarity with basic machine learning concepts — training and validation splits, class balance, inter-rater agreement Experience with dataset quality control or annotation review workflows Comfort working alongside engineers and clinicians to resolve labeling questions Pay: $20.00 - $25.00 per hour Expected hours: 8.0 – 40.0 per week Work Location: In person