Data Annotation Specialist

Marionet - San Francisco, CA

Marionet is looking to hire a Data Annotation Specialist to work on-site at their office (cool space, with free drinks and snacks).

Marionet is a seed stage startup, building a robot to bring automation, affordability and convenience to the beauty services market.

As Data Annotation Specialist, you will take on data annotation and validation tasks to help train machine learning algorithms. You will use a variety of internal and external software programs to mark objects, or other items of interest in images.

The ideal candidate has great attention to detail, is comfortable doing repetitive work with data, and will be responsible for helping annotate large datasets for algorithm training. Duties also require providing feedback for improvements on tools and processes, and on occasion, helping field teams gather data onsite.


  • Annotate and label image and video data via computer using various internal software programs.
  • Attach metadata (comments, markups, etc.) to image data in accordance with established procedures.
  • Provide constructive feedback to our engineers on tooling and processes for more efficient labeling, and to improve data quality.
  • Maintain ownership of each annotation assignment, and respond to any changes in procedure and auditors.
  • Evaluate metrics to help improve our productivity.
  • Communicate clearly using excellent written and verbal skills.
  • Meticulous, detail oriented, and patient.
  • Must be reliable, proactive, committed and quality focused.
  • Able to maintain sharp focus, even with repetitive tasks.
  • Willing to support several projects at one time, and to accept reprioritization as necessary.
  • Some experience of similar activities would be a bonus, e.g. image segmentation and classification, data entry, etc.
Attention - In the recruitment process, legitimate companies never withdraw fees from candidates. If there are companies that attract interview fees, tests, ticket reservations, etc. it is better to avoid it because there are indications of fraud. If you see something suspicious please contact us: [email protected]