V2 endpoint — requires Pro, Team, or Enterprise plan.
async_index: true, each artifact gets a job_id. Use this endpoint to check whether indexing has completed and the artifact is ready to retrieve.
Request
Response
Current state of the indexing job.| Value | Meaning |
|---|
pending | Queued, not yet started. |
processing | Indexing in progress. |
done | Complete. Artifact is available for retrieval. |
failed | Indexing failed. See error_message. |
UUID of the artifact. Present once status is done.
Workspace this job belongs to.
ISO 8601 timestamp of the most recent status change.
Human-readable error if status is failed. null otherwise.
Example
import time
from nined.memory import MemoryClientV2
client = MemoryClientV2(api_key="your-key", workspace_id="ws_ops")
result = client.ingest(
[{"artifact_type": "document", "raw_payload": {"content": large_doc}}],
async_index=True,
)
job_id = result["queued_jobs"][0]["job_id"]
# Poll with backoff
for _ in range(20):
status = client.job_status(job_id)
if status["status"] == "done":
print(f"Ready: {status['artifact_id']}")
break
if status["status"] == "failed":
print(f"Failed: {status['error_message']}")
break
time.sleep(2)
{
"job_id": "j1a2b3c4-...",
"workspace_id": "ws_ops",
"status": "done",
"artifact_id": "a1b2c3d4-...",
"created_at": "2025-03-30T12:00:00Z",
"updated_at": "2025-03-30T12:00:03Z",
"error_message": null
}
For most indexing jobs, done status arrives in under a few seconds. Poll every 2 seconds with a backoff cap at 30 seconds for large documents.
