Healthcare receivables agency
≈ $10M revenue (approx.) · ~60 employees · national client base
Placement-file intake automation
Placement files arriving in any layout are normalized into the agency's existing upload format, with a human approval queue before anything reaches the system of record. A one-and-a-half-person manual job becomes a review step.
- Organization
- Healthcare receivables agency, on-premise, regulated
- Volume
- Placement files in twenty-five to fifty column layouts, from many clients
- Runs on
- The agency's own hardware
- Scope
- Intake normalization with a human approval queue
Where they started.
Every client sends placement files in its own layout. A person and a half spent their days turning those files into the one upload format the thirty-year-old collection system accepts. A failed system conversion the year before had made the agency rightly cautious: whatever came next could not touch the system of record.
What was built.
- The agent watches the drop locations, reads each incoming placement file, and maps its columns onto the house schema using rules the intake team helped write.
- Anything it is unsure about is flagged, not guessed; the whole batch waits in an approval queue where a reviewer sees the mapping and the exceptions before release.
- Output is exactly the upload file the collection system already accepts, so nothing about the system of record changes.
- Runs entirely on hardware inside the agency's building.
What changed.
- Intake becomes a review step instead of a data-entry job.
- New client layouts are handled by adding a mapping, not by retraining a person.
- Zero changes to the legacy system that the business depends on.
- Every release is auditable: who approved which batch, and what was flagged.
Placement files carry protected health information. The agency's compliance posture depends on that data never leaving the building, so the agent runs where the data already lives.
On-premise hardware
Components involved.
More like this.
Healthcare receivables agency
≈ $10M revenue (approx.) · ~60 employees · regulated healthcare receivables
- Use Case:
- Protected-health-information send gate
- Results:
- Every outbound file that carries protected health information is checked against the placement it belongs to and held for human approval, with an audit trail of who released what, and when.
Sleep-therapy provider
≈ $60M revenue (approx.) · 40 clinics · one clinical reporting app
- Use Case:
- Natural-language clinical reports
- Results:
- Clinicians ask for patient cohorts in plain English and a constrained pipeline turns the question into validated filters, never free-form database queries, on a private GPU that costs a fraction of the hosted model it replaced.
Global law firm
≈ $3B+ revenue · Thousands of lawyers · dozens of offices
- Use Case:
- Walled-matter research assistant
- Results:
- Document search and drafting inside the firm's own tenancy: every answer cited to its source, every matter isolated by role, and nothing ever sent to a third-party model.
Have one like it?
Every spotlight started as a conversation about a process nobody liked doing, under a data-locality constraint.