What it does
An intake system for a pharma drug-safety team. Reports arrive as emails, call-centre notes, web forms and scanned letters. Each one is read, the patient, product, event and reporter are pulled out, duplicates are checked, and the case is ranked by seriousness and due date. A safety reviewer confirms every case before it is logged; incomplete ones go back out for follow-up.
The problem
Reports arrive in every format and each one has a regulatory clock. Entering one by hand takes around 15 minutes, and a serious case can sit in a shared inbox behind routine ones.
Approach
AI reads each report and fills a case draft, marking how sure it is about every field. Rules check the minimum details, look for duplicates and seriousness terms, and set the due date. Nothing is submitted on its own: a reviewer accepts, edits or returns each case, and every change is kept in an audit log.
- Input
- AI step
- Rules or code
- Person decides
- Result
Screens
Swipe sideways to see the whole screen.
Client work is shown under confidentiality agreements: no client names, and screens are recreated with sample data.
Stack
- Gemini API
- OCR
- Rule engine
- Case tracker with audit log
Impact
Typical for systems like this.
