AI Automation
Clawleaf: Data Extraction & Reporting
Ten bots for recurring reports, bulk record retrieval, mileage, outcomes and change alerts.
DisciplinesAI Automation
Our roleDesign, build, release
Technologies
Python
PyTorchVision modelsInference at edge
PythonPyTorchVision modelsInference at edge
Clawleaf
The challenge
What was in the way.
Operational reporting was manual and therefore late: census counts, new-patient digests, provider mileage and discharge outcomes were all assembled by hand, alongside one-off bulk record pulls that took days of clicking.
Disciplines in scope
Hardware—
Firmware / IoT—
AI / MLIn scope
Web & mobile—
How we built it.
Recurring reports and one-time extractions share the same extraction core, so a bulk pull built for a migration becomes a scheduled report with a config change. Each run is idempotent and leaves an audit trail of what it read and produced, and the reports are delivered where the work already happens — email and team channels — rather than a dashboard nobody opens.
01Technical discovery
02Architecture & prototype
03Build & iterate
04Deploy & hand over
Weekly Census Reporter — counts patients per facility and per provider and emails a weekly report
Daily New-Patient Digest — sends the physician a summary of new patients each day
Daily Mileage Calculator — calculates provider travel mileage each day from visit data
Discharge-Outcome Tracker — looks back over recent patients to capture where each one went after discharge
Bulk Data Export Bot — one-time bulk export of records from a practice-management system
Medical Records Retrieval — one-time bulk retrieval of patient medical records
Monthly List Processor — processes a recurring monthly worklist
AI Hospitalization Review — uses AI to review hospitalization records
Records Portal Bot — automates a medical-records portal to retrieve and process records
File Change Monitor — watches a shared spreadsheet and alerts the team the moment it changes
01Shipped
Delivered to production and still maintained by the team that built it.02Tested
One test matrix across hardware, firmware and interface — no gaps between layers.03Documented
Runbooks, monitoring and escalation handed over with the system.04Scaled
Grew with usage without re-architecture or a second rebuild.