Generalist for Data Annotation
RemoteContract / Freelance
12 open roles · Updated
Before a model can recognize anything, someone has to label it. Boxes drawn around objects in a video, speakers separated in an audio file, categories assigned to images, text marked up so a system can learn the pattern. That labeling is data annotation, and it is the most common entry point into this industry.
The appeal is that most roles need no degree and no prior experience. The part people underestimate is that annotation is exacting work. Each project ships with guidelines that define exactly how to label an edge case, and a batch where the guidelines were applied loosely is worth nothing to the lab that ordered it. Quality review is continuous, and consistency is the thing being measured.
Pay here is modest, which is the honest consequence of it being the work with the fewest barriers to entry. It is a reasonable way in, and people who stay in the industry usually move toward evaluation or a specialism within a year.
If you are weighing whether this is worth your time, we have written about whether data annotation is legitimate work and what the roles and requirements actually are.
RemoteContract / Freelance
RemoteContract / FreelanceMid-level
RemoteContract / Freelance
RemoteContract / Freelance
RemoteContract / FreelanceMid-level
RemoteContract / FreelanceAdvanced / Expert
RemoteContract / Freelance
RemoteContract / Freelance
RemoteContract / Freelance
RemoteContract / Freelance
RemoteContract / FreelanceAdvanced / Expert
RemoteContract / FreelanceAdvanced / Expert
If a company stops meeting these standards after being listed, its roles come off the site.