Payment integrity
Knowledge Center
Centene · payment integrity
Dataset
Payment integrity policy
The internal payment policies the claims in this batch are tested against, as source documents.
Documents
7
18 pages ingested
Chunks indexed
32
1000 characters, overlapping by 200
Embedding model
local-hashing-256
256 dimensions · recorded at first ingest
Ingestion runs
7
7 completed
How this dataset is indexed
Organisation
org-centene-payment-integrity
Project
prj-claims-audit-2026
Source
Local filesystem (the platform uses S3, Azure Blob, GCS or SharePoint)
Embedding, here
local-hashing-256 at
256 dimensions —
a local deterministic hashing embedding, not a learned model
Embedding, on the platform
text-embedding-3-small at
1536 dimensions via
Azure OpenAI, or amazon.titan-embed-text-v2:0 at 1024 via Bedrock
The dimensions differ, so a dataset cannot be moved between the two without a fresh collection and a re-ingest — which is true of the platform's own providers as well.
Documents
| Document | Policy | Version indexed as | Pages | Chunks | Digest |
|---|---|---|---|---|---|
|
CPP-061_physician-administered-drugs_v2.0.md
Seeded with the demo · policy committee, 12 Feb 2026 · 18 Sep 2026 18:47 UTC
|
CPP-061
v2.0
|
__live__
in force
|
3 | 6 | 9bb370474c9d… |
|
CPP-061_physician-administered-drugs_v1.4.md
Seeded with the demo · superseded, retained for claims before 01 Mar 2026 · 18 Sep 2026 18:47 UTC
|
CPP-061
v1.4
|
v1.4
superseded, still retrievable
|
2 | 3 | 648c822e0c9d… |
|
CPP-047_inpatient-drg-validation_v3.1.md
Seeded with the demo · facility audit policy, 04 Dec 2025 · 18 Sep 2026 18:47 UTC
|
CPP-047
v3.1
|
__live__
in force
|
3 | 5 | 3f7229f9774e… |
|
CPP-085_prior-authorization-post-service-review_v3.4.md
Seeded with the demo · utilization management, 09 Oct 2025 · 18 Sep 2026 18:47 UTC
|
CPP-085
v3.4
|
__live__
in force
|
3 | 6 | 445dfcdd4825… |
|
CPP-080_eligibility-enrolment-retroactive-termination_v6.1.md
Seeded with the demo · enrolment integrity, 18 Dec 2025 · 18 Sep 2026 18:47 UTC
|
CPP-080
v6.1
|
__live__
in force
|
3 | 5 | 68f9cca62e0a… |
|
CPP-014_em-level-selection_v4.2.md
Seeded with the demo · policy committee, 21 Nov 2025 · 18 Sep 2026 18:47 UTC
|
CPP-014
v4.2
|
__live__
in force
|
3 | 4 | 37f66b14d2c6… |
|
CPP-021_duplicate-and-repeat-services_v2.3.md
Seeded with the demo · claims editing governance, 12 Jun 2025 · 18 Sep 2026 18:47 UTC
|
CPP-021
v2.3
|
__live__
in force
|
1 | 3 | b193fd20e1e9… |
Ingestion runs
| Run | Document | Status | Read | Pages | Chunks | Embedded | Stored | Errors |
|---|---|---|---|---|---|---|---|---|
| run-0007 | CPP-021_duplicate-and-repeat-services_v2.3.md | completed | 1 | 1 | 3 | 3 | 3 | — |
| run-0006 | CPP-014_em-level-selection_v4.2.md | completed | 1 | 3 | 4 | 4 | 4 | — |
| run-0005 | CPP-080_eligibility-enrolment-retroactive-termination_v6.1.md | completed | 1 | 3 | 5 | 5 | 5 | — |
| run-0004 | CPP-085_prior-authorization-post-service-review_v3.4.md | completed | 1 | 3 | 6 | 6 | 6 | — |
| run-0003 | CPP-047_inpatient-drg-validation_v3.1.md | completed | 1 | 3 | 5 | 5 | 5 | — |
| run-0002 | CPP-061_physician-administered-drugs_v1.4.md | completed | 1 | 2 | 3 | 3 | 3 | — |
| run-0001 | CPP-061_physician-administered-drugs_v2.0.md | completed | 1 | 3 | 6 | 6 | 6 | — |
These are the counters the platform's own document_ingestion
workflow emits, with errors bucketed by stage — a read failure and an
embedding failure are different problems for whoever has to fix them.
The pipeline that ran
storage_list_files
List the files in the dataset's source path
document_reader_unified
Read each file — PDF, DOCX, XLSX, PPTX, HTML on the platform; .txt and .md here
metadata_extractor
Pull filename, page count, digest and the policy the document belongs to
text_splitter_v2
Split each page into 1000-character chunks overlapping by 200
batch_embedder
Embed every chunk — local-hashing-256 at 256d here, text-embedding-3-small at 1536d on the platform
zip_merge
Join each embedding back to its chunk and its metadata
batch_vector_upsert
Write the points, each stamped with organization, project, dataset, document, version and is_active