Frequently Asked Questions

Pricing depends on data modality, task complexity, QA depth, and volume. We provide custom estimates after understanding your requirements. Typical engagements start with a Pilot Pod ($2,500–$7,500).

No long-term contracts required. Our Pilot Pod is designed as a low-risk starting point. You can scale to a Dedicated Pod or Burst Capacity based on results.

Yes. Our Pilot Pod (10–14 days) lets you evaluate quality, speed, and communication before committing to an ongoing engagement.

We use AES-256 encryption at rest, TLS 1.3 in transit, role-based access control, and workspace-level data isolation. We execute NDAs and DPAs as standard.

We follow SOC 2-ready processes and controls. Formal certification is in progress. We’re happy to share our security practices in detail during evaluation.

Yes. We offer HIPAA-aligned workflows with access controls, audit trails, and de-identification support for healthcare data.

Accuracy targets are set per project and depend on task complexity. We typically achieve 95–99%+ accuracy with our multi-layer QA pipeline. Specific SLAs are defined during project scoping.

We use inter-annotator agreement (IAA) metrics including Cohen’s Kappa, Fleiss’ Kappa, and IoU scores. These are tracked per project and reported weekly.

We build edge-case libraries during guideline creation, maintain a disagreement taxonomy, and use structured adjudication protocols for ambiguous cases.

Depends on volume and complexity. Pilot Pods deliver in 10–14 days. Dedicated Pods typically process 10K–100K+ labels per week at steady state.

We’re tool-agnostic and work with major platforms (Labelbox, CVAT, Label Studio, Prodigy, custom tools). We can also integrate with your existing tooling.

We support JSON, JSONL, CSV, COCO, Pascal VOC, YOLO, and custom formats. We match your ML pipeline’s requirements.