A self-service data labelling platform.

Take control of your data labelling projects and leave all the annotating to us.

View pricing

“The annotations finished overnight, which was amazing. Speed is important to us. It's not something we see from many other platforms."

Jacob Rogers,
Software Engineer

We deliver ease, speed and quality.

1

Easy to use

Streamline your projects by following our simple step-by-step process to upload images, create instructions, evaluate quality and export labelled data.

2

Quick results

Complete your projects at lightning bolt speed. A Databolt user once completed 5 different projects with 2,500 annotations in total - in less than 2 hours!

3

High quality

Quickly iterate and improve your projects by viewing annotated tasks, evaluating quality, and updating labelling instructions.

GIF illustrating rhe 3 steps it takes to set up a project in Databolt.

How it works

You manage, we annotate.
‍

1. Set up a project

Upload images, add labels and set instructions. Start when you're ready!

2. Review annotations

Review annotated tasks, assess annotation progress and improve labelling instructions as needed.

3. Export data

Once the annotation tasks are complete, export the labelled data and start training your ML models.

Pricing

Bounding box

USD

$0.07

per annotation

Try now

Polygon

USD

$0.18

per annotation

Try now

Polyline, point and semantic segmentation

Coming soon

I'm interested!

A seamless experience at each touchpoint.

Pay as you go

No platform fees, upfront commitment or minimum spend required. We’ll only charge your credit card once the data is labelled.

Create instructions

Our ready-made instruction templates will help you craft clear and precise guidelines for our labellers. Choose a template based on your project type and get the results you need.

Review progress

View each annotated task and assess the quality as soon as it is complete. Be quick to iterate and improve your data labelling projects based on these in-depth insights.

Evaluate quality

With our comprehensive dashboard, users are able to assess the overall precision score, mislabelled tasks, false positives and other in-depth data to evaluate quality.

Got a question?

What annotation types are available?

Currently, we only support bounding box and polygon annotation types. Polyline, point and semantic segmentation are coming soon.

What output format is supported?

Currently, we only support COCO JSON output format. We'll be introducing other output formats soon. If you require another output formats, drop us a message at support@databolt.ai.

How quickly can I expect to receive annotated images?

Our annotators will start labelling your data as soon you click Start Project. Turnaround time will depend on the complexity of the project. One of our early users received annotated images with around 2,500 annotations in less than 2 hours.

How do you ensure high quality annotation output?

Our instructions templates will help you provide clear labelling guidelines so our annotators understand how to accurately label your images. You can then use Databolt's quality evaluation tools to identify common errors and iterate accordingly by improving the instructions or label definitions.

Who will label my data?

Our annotators will label your data after being vetted through a series of training and assessments before being onboarded to bolt. They are remote annotators who work part time on different projects available on our platform. Annotators are assigned automatically to projects on the platform.

What payment methods are available?

Currently, we only accept payments via credit card.

GIF depicting the quality tools available in Databolt

Get labelled data in hours, not days.

Start project

Terms & Privacy