Data Annotation Specialist - Computer Vision

Bobyard•San Francisco, CA

1d•$20 - $35•Remote

About The Position

Bobyard uses computer vision and NLP to automate takeoffs and estimates in the construction industry. This role involves labeling construction drawings to train computer vision models. The work requires meticulous attention to detail and consistency to ensure the accuracy of the data used by ML engineers. As the company expands into new trades, the need for labeled drawings increases significantly. The position is described as repetitive and exacting, with the expectation that labels will be of high quality and consistency.

Requirements

Ability to identify errors, inconsistencies, or omissions proactively.
Skill in precisely following written instructions and applying good judgment when instructions are insufficient.
Capacity for focused, repetitive work over extended periods without compromising accuracy.
Ability to learn new tools quickly and work independently in a remote setting.
Clear written English for issue reporting and communication.

Nice To Haves

Prior experience with image annotation tools such as Labelbox, CVAT, or Supervisely.
Basic understanding of computer vision concepts like object detection and segmentation.
Ability to read construction, CAD, or architectural drawings.

Responsibilities

Label construction drawings with bounding boxes, polygons, and segmentation masks that meet accuracy standards.
Apply guidelines consistently to ensure uniformity in labels.
Identify and flag ambiguous or edge cases, and contribute to refining guidelines.
Clean existing datasets by correcting inconsistent labels, missing metadata, and duplicates.
Meet weekly targets for quality and throughput.
Communicate observations about the data to ML engineers to expedite issue resolution.