Hire an annotation team for the tools you use

Label vehicles in a sample image batch inside your approved Ground Truth workforce setup. Ground Truth is closed to new customers. Existing customers can request this pilot in their own workforce setup.

Mark invoice numbers, dates and totals in a small set of synthetic documents in your Ango Hub project.

Fit 3D cuboids to parked vehicles in a short LiDAR sequence from your BasicAI workspace.

Label an image sequence and collect cases where objects disappear or the class list is incomplete.

Draw boxes around visible surface indications in sample inspection images using the Diffgram version installed in your environment.

Have two reviewers independently label a sample of visually similar defects in your Kili workspace.

Mark complaint categories and evidence spans in a small set of anonymized support messages.

Compare two possible answers for each prompt in a small evaluation set with the same rubric.

Review a pilot batch against your existing Scale Studio taxonomy and record ambiguous examples for adjudication.

Trace the edges of small components in close-up product images using your class definitions.

Track moving equipment in a short video using persistent object identifiers.

Label a set of components at different scales in your available Datature workspace.

Mark the start and end of events in a short video sequence against your event definitions.

Box packaged products on shelf images, including partially hidden items.

Track one object through a short sequence of synchronized camera and LiDAR frames in your Segments.ai project.

Distinguish touching instances of the same class in a sample of inspection images.

Trace the outline of one part across consecutive inspection images using the client’s agreed classes.

Rank pairs of assistant responses and explain disputed choices using the reference answer.

Mark organization names and contract dates in synthetic procurement text.

Classify customer intents and mark the phrases that support each decision in a sample corpus.

Review named-entity spans in your existing LightTag workspace or an approved export of its annotations.

Review proposed entity spans sentence by sentence using a task recipe you approve.

Independently review records where your labeling rules disagree or abstain.

Trace land-use boundaries on a sample of georeferenced imagery.

Run a trial in a restricted test project using your internal interface and written instructions.