The three APIs the DataOps side implements

When you register one base URL with the platform, the platform calls three paths under it. In the examples, https://data.example.com/api is the base URL.

The paths must have exactly this shape. In return, the field names in responses can follow your own conventions — the platform accepts several names with the same meaning (see Field aliases below).

Common conventions#

1. Dataset list#

curl -H "Authorization: Bearer $TOKEN" \
  "https://data.example.com/api/datasets?page=1&page_size=50&updated_after=2026-07-01T00:00:00Z"

{
  "items": [
    {
      "id": "ds-inspect-2026w29",
      "name": "Line inspection images 2026-W29",
      "modality": "image",
      "file_count": 412,
      "size_bytes": 1073741824,
      "updated_at": "2026-07-21T09:30:00Z"
    }
  ],
  "page": 1,
  "total": 3
}

• The platform sends page (starting at 1) and page_size. If you do not support them, you may ignore them and return everything. If you supply total, the screen uses it to decide on "Load more".
• If updated_after is given, return only what changed after that time (recommended).
• The connection test calls once with page=1&page_size=1.

2. Manifest#

curl -H "Authorization: Bearer $TOKEN" \
  "https://data.example.com/api/datasets/ds-inspect-2026w29/manifest"

{
  "dataset_id": "ds-inspect-2026w29",
  "annotation_format": "coco",
  "classes": [{ "id": 0, "name": "defect_a" }, { "id": 1, "name": "defect_b" }],
  "files": [
    {
      "file_id": "f-0001",
      "filename": "20260721_line3_0001.png",
      "kind": "image",
      "size_bytes": 2493833,
      "sha256": "9f8a1c…",
      "meta": { "line": "line-3", "measured_at": "2026-07-21T09:30:12+09:00" }
    }
  ]
}

3. File body#

curl -H "Authorization: Bearer $TOKEN" -o 0001.png \
  "https://data.example.com/api/datasets/ds-inspect-2026w29/files/f-0001"

• Stream the file body as is. Any Content-Type works.
• Redirects are followed, so you may return a 302 to a presigned URL instead of the body.
• Resuming (Range) is not used yet. A failed file is fetched again in full on the next sync.
• Each file can be up to 2 GiB (configurable in the server settings).

Error responses#

// 401
{ "error": "invalid_token", "message": "The token has expired." }
// 404
{ "error": "dataset_not_found", "message": "ds-xxxx was not found." }
// 429 (sending a Retry-After header along with it is helpful)
{ "error": "rate_limited", "message": "Please try again later." }

The platform does not interpret this body; it shows the start of it to the user. Put in one sentence that reveals the cause.

Field aliases#

You can use any of the names below for response fields. The platform looks for them in order and uses the first non-empty value.

If you really need a name not in this table, tell the platform team. It can be accepted by adding one line.

Supported file formats#

CSV · Parquet files cannot be told apart as time series or tables by extension, so either report modality as tabular in the list, or have the modality chosen directly when the dataset is created.

Integration checklist (provider side)#