M
Medigy
03 / Ingestion
Book a demo →

From raw files to structured assets.

Ingestion is file-based, API-aware, and batch-oriented — built around repeatable jobs, not fragile glue code.

For:BD leadersProductData teams

How data gets in

  • — File-based intake for CSV, Parquet, JSON, and XML
  • — API-based pulls where source endpoints are available
  • — Batch-oriented with idempotent, repeatable jobs

Surveillance, not ETL

The Medigy Opportunity Atlas uses a data surveillance approach to track every dataset from source to transformation. Each dataset is versioned, every source is provenance-tracked, every transformation is logged, and lineage is queryable at any point.

  • — Every dataset is versioned
  • — Source provenance is tracked (URL, hash, fetched-at)
  • — Transformations are logged and reproducible
  • — Data lineage is visible and queryable

Three-stage flow

Stage 1
Raw

As-is from source. Bytes preserved, file hashes recorded.

Stage 2
Staged

Cleaned, typed, deduplicated. Source semantics intact.

Stage 3
Canonical

Normalized into the Opportunity Atlas model. Joinable across sources.

What this means for you
If a regulator, partner, or your own team asks "where did this number come from?" — you can answer with the exact source file, version, and SQL that produced it.
See this in your data

Ready to put the Opportunity Atlas to work?

See the Opportunity Atlas run against your product, segment, or geography in a 30-minute walkthrough.