Architectural Floor Plan Segmentation Labeler

About OpenTrain

OpenTrain is the #1 platform for finding and building careers in AI training and data labeling. We connect people to hands-on projects that teach AI how the world looks, sounds, and reads.

OpenTrain AI is the hiring and contracting organization for this role; contributors are engaged directly on contract to help train production-ready models.

About AI Training Work

AI training (data labeling/annotation) is the human side of building intelligent systems. Labelers create the example data modern models learn from by drawing masks, tagging objects, transcribing audio, and rating outputs.

This work is often remote, flexible, and accessible: you don’t need to be a professional developer to participate, but accuracy and consistency are essential.

The Role

You will annotate architectural floor-plan images so a segmentation model can detect room boundaries and classify occupancy types. Tasks are done in Roboflow using polygon and segmentation tools.

This is a part-time contractor role. Expected commitment is 20+ hours per week. Pay is hourly at USD $10/hr. The project requires intermediate labeling experience and strong attention to detail.

What You'll Do

• Draw polygon masks around each room shown in a floor plan, accurately following printed walls and boundaries.
• Label each polygon with the correct occupancy type (e.g., Bedroom, Bathroom, Corridor, Storage, Mechanical, Office, etc.).
• Ensure polygons are clean, non-overlapping, and precisely aligned to room edges.
• Verify that any visible text labels or abbreviations match the assigned occupancy type and correct labeling when needed.
• Annotate an average of 15–20+ individual items per file across residential, commercial, and mixed-use plans.
• Work in Roboflow using polygon and segmentation tools and follow written instructions and examples for consistency.

Requirements

You must have hands-on experience with polygon/segmentation labeling in Roboflow or comparable tools (CVAT, Labelbox, etc.).

Ability to read and interpret basic architectural drawings — identifying walls, room boundaries, doors, and common abbreviations — is required.

• Prior experience with Roboflow, CVAT, Labelbox, or similar annotation platforms.
• Ability to read and interpret basic floor-plan conventions (walls, doors, text labels, abbreviations).
• Strong attention to detail and consistency when drawing masks and assigning room types.
• Comfortable following detailed written instructions and using examples to match labeling standards.
• Reliable communication and ability to meet deadlines for batch-based work.
• Preferred (not required): background in architecture, drafting, civil engineering, or a related field.

How Work Is Structured & Tools

Source files are high-resolution raster images extracted from architectural PDFs. Floor plans vary widely in style, clarity, and density.

Work is delivered in batches; each file contains many rooms and typically requires 15–20+ labeled polygons. Detailed instructions and examples will be provided to help you meet quality standards.

• Label types used: bounding boxes, polygons, and segmentation masks.
• Primary tooling: Roboflow (polygon and segmentation tools).
• Work is contract-based and remote; you can contribute from anywhere worldwide.

Compensation & How to Apply

Compensation is USD $10 per hour. This is part-time contractor work with an expected commitment of 20+ hours per week.

To apply, create or sign in to your OpenTrain account and submit your profile highlighting relevant labeling experience and any architectural/drafting background. OpenTrain AI will contract contributors directly.

• Employment type: Contractor, part-time.
• Worldwide applicants welcome; language requirements are not specified here.
• Applications should demonstrate prior annotation tool experience and familiarity with floor plans when possible.