Prédilis
« L'IA prépare. Vous soignez. » AI prepares. You care.
A platform that structures preventive health check-ups, so care professionals spend the consultation on the patient instead of the paperwork. I lead its Django development at Predilib.
- Role
- Lead Django Developer
- Company
- Predilib, France
- Period
- Oct 2025 – present
- Stack
- Django · DRF · PostgreSQL · Signals
Figures published on predilis.fr, September 2026.
Prepared before the patient walks in.
Before the appointment, the patient fills in a questionnaire. Prédilis turns the answers into an intake summary and a pre-filled consultation framework, and flags risks automatically. The practitioner leads the consultation and validates every finding. The platform is an organisational aid, not a medical device.
Data is hosted in France on HDS-certified infrastructure (the French standard for health-data hosting) and handled in line with GDPR, with a complete audit trail.
- General prevention check-up, for every eligible profession
- Shared medication review (BPM), for pharmacists
- Asthma and anticoagulant (AVK/AOD) follow-up interviews
- Pregnancy support (EFE) and opioid management (EPA)
How the back end fits together
A simplified view of the parts I work on. The request path runs along the top; background work runs underneath.
What I built
As lead Django developer, I own the back end from database design to the endpoints the front end calls.
Built the platform's back-end infrastructure in Django as a modular system that scales and stays maintainable.
Designed and optimised the PostgreSQL databases through the Django ORM, for secure and efficient storage of sensitive patient data.
Built RESTful APIs with Django REST Framework, giving the front end and external services secure endpoints.
Integrated several third-party APIs to sync patient data automatically and enrich health reports, handling authentication, errors and data transformation.
Used Django Signals to trigger follow-up work automatically when data changes, making interactions feel real-time and report generation more efficient.
Built a responsive front end with HTML, CSS, JavaScript and AJAX, with smooth interactions and dynamic updates on every device.
Applied role-based access control, multi-factor authentication and data encryption, in line with health-data standards.
Added caching and query optimisation to cut response times and make the experience faster.
Related note: Django signals: where side effects belong →