Tools & Delivery
Work in agreed tools. Deliver in usable formats.
Omni Annotate AI aims to fit into your existing data workflow where practical. Before accepting a project, we confirm the tools, data access, output format, quality requirements and delivery schedule. We only commit to capabilities we can support for that specific engagement.
Working with your platform
Where compatible, we work in a client provided annotation or evaluation platform with authorized access. Another tool or file sharing workflow may be agreed when appropriate. Platform access and task feasibility are checked before project data is transferred, particularly for image, video, audio and LiDAR assignments.
Example delivery formats
The following are illustrative options, not an unconditional promise of compatibility. The exact schema, tool support and handoff method are confirmed during scoping.
| Work type | Possible outputs, subject to project compatibility |
|---|---|
| Text annotation and structured data | CSV, JSON, JSONL, spreadsheets or client defined schemas |
| Document annotation and OCR | Corrected text, structured fields, document labels and agreed extraction records |
| Image annotation | Tool compatible COCO, YOLO, Pascal VOC or other agreed annotation exports |
| Video annotation | Frame level labels, timestamps, tracking records or platform specific exports |
| Audio and transcription | Transcripts, speaker/segment labels, timestamps, SRT or WebVTT where supported |
| LiDAR and 3D point clouds | Supported 3D object labels and point cloud annotation structures; advanced exports are scoped separately |
| AI evaluation | Rating sheets, structured scores, error logs, reviewer notes and evaluation summaries |
Defining the project before work starts
We agree on the task, input data, sample or pilot approach, output structure, number of items, acceptance criteria, review depth, price and timeline. For small, well defined tasks, the first agreed batch can serve as calibration instead of a separate pilot.
Milestones and delivery
Work may be organized by project, batch or milestone. The delivery schedule is confirmed after requirements, data readiness, tool access and capacity are reviewed. Each delivery includes the agreed quality summary or report, reflecting checks actually performed.
Changes and version control
If the client changes label definitions, scoring rubrics or output requirements, we review the effect on completed work, volume, price and deadlines. Updated instructions and reference examples are documented, and any scope changes are approved before implementation.
Acceptance and corrections
Deliverables are reviewed against the agreed acceptance criteria. In scope corrections are handled during the agreed review period. Additional tasks or new specifications are separately discussed and approved.
Working alongside an existing provider
Where the client authorizes it and the assignment fits our capacity, we can support a defined portion of an existing workflow, such as a scoped annotation task, sample review, transcription check or selected error analysis. This is assessed case by case.
What we need to prepare a quotation
A short description of the task, data type, approximate volume, sample where safe to share, preferred output format, relevant guidelines and target schedule. Confidential files can be shared after appropriate arrangements are in place.
Start With a Project Discussion
Tell us about your data, project objectives, and the annotation, training data preparation, or AI evaluation support you need. A brief description is enough to get started.
We will review your requirements, assess project feasibility, and discuss suitable next steps. Where the project aligns with our capabilities, we can prepare a tailored proposal covering scope, pricing, deliverables, timelines, and acceptance criteria.