AI Automation
Clawleaf: Signature & Approval Automation
Nine bots that clear signature and approval queues across every facility, every day.
DisciplinesAI Automation
Our roleDesign, build, release
Technologies
Python
PyTorchVision modelsInference at edge
PythonPyTorchVision modelsInference at edge
Clawleaf
The challenge
What was in the way.
Clinicians were spending hours a day on signatures — orders, certifications, clinical notes and e-signature task lists piling up across separate systems, each with its own login and its own queue.
Disciplines in scope
Hardware—
Firmware / IoT—
AI / MLIn scope
Web & mobile—
How we built it.
Each queue was mapped as a state machine before a line of automation was written: what counts as signable, what must never be auto-signed, and where a human has to step in. The bots share one credential vault and one-time-passcode handler, run on a daily schedule, and stop and escalate rather than guess whenever a document falls outside its rule set.
01Technical discovery
02Architecture & prototype
03Build & iterate
04Deploy & hand over
Order Signature Bot — signs every order awaiting signature across all facilities, clearing the queue daily
Order-Review Clearing Bot — works through pending order-review queues so approvals never pile up
Certification Signature Bot — applies the physician's signature to recurring patient-care certifications in bulk
Clinical Note Signing Bot — opens and signs each pending draft clinical note in sequence
Bulk E-Signature Bot — clears an e-signature task list, signing each item with a secure passphrase
One-Click Bulk Signing — accepts every pending item, then signs them all with a single password entry
Secure Sign-In & Signature Bot — logs in via one-time passcodes from email and signs prepared documents, with safeguards when extra verification is required
Care-Document Signature Bot — applies signatures to recurring care and care-task documents
Hospital Order Signature Bot — signs pending orders inside a partner hospital's system
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.