Oslerfy —A free skillshare app but for resident doctors in the hospitals
Inspiration
In hospitals, procedures happen every day that resident doctors need to learn. But the opportunity often passes unnoticed.
A consultant might think this would have been great for a resident doctor, but there is no way to signal it quickly. Meanwhile a resident doctor finishes a shift having missed another chance to practise a required skill.
Across the UK NHS, 62,000 resident doctors must build verified procedural portfolios to progress. The issue is not lack of opportunity. Hospitals are full of procedures. The issue is visibility and coordination.
Oslerfy exists to make those moments visible and to connect teachers and learners at the moment the skill is being performed.
What It Does
Oslerfy is an iOS app for real-time procedural learning.
A teacher can broadcast that a procedure is about to happen. Resident doctors nearby see the opportunity and request to join. If accepted, they can coordinate before the case begins.
Afterwards, both participants complete a dual digital sign-off. The teacher confirms the resident doctor’s participation and performance. The learner records the experience, reflects on the skill (optional), and gives feedback to the teacher as well, because teachers also need to demonstrate evidence of teaching during their performance reviews.
Both sides benefit. The resident doctor gains verified training evidence. The teacher records teaching activity and contributes to peer learning.
The full loop is simple:
- Discover – a teacher posts a live procedure opportunity.
- Join – a resident doctor requests to participate.
- Learn – both coordinate before and during the session.
- Verify – teacher and learner complete a mutual sign-off, beneficial to both.
- Reflect – the resident doctor records notes and generates portfolio evidence.
This turns real clinical work into structured training without extra admin.
How I Built It
Frontend SwiftUI iOS application with dedicated ViewModels and a live opportunity feed.
Backend Supabase with Postgres for users, sessions, and logbook data. Realtime subscriptions power the live feed and row-level security isolates user data.
AI features Used in four focused places: procedure briefings, reflection prompts, learning plan suggestions, and portfolio summaries. All requests pass through Supabase Edge Functions and are anonymised before being sent.
Services Session management, messaging, push notifications, location context, PDF portfolio export, CSV export, and optional in-app coffee gifts for teachers.
Challenges
Clinical data privacy Any training data leaving the device required anonymisation. Identifiers are stripped before external processing. Geofencing to specific hospital sites.
Keeping features focused It was easy to add more automation and complexity. Anything that did not remove a real friction point in the workflow was cut.
Designing the dual sign-off flow Verification needed to be fast and natural on a phone, often done between patients.
Accomplishments
- A working end-to-end loop from opportunity broadcast to verified portfolio evidence.
- A dual sign-off model that recognises both the teacher and the learner.
- Real-time coordination of teaching moments inside a hospital.
- A simple interface usable mid-shift with no onboarding.
What I Learned
The biggest challenge is over-engineering a simple solution.
Doctors operate in fast, high-pressure environments. If recording a teaching moment takes more than a few seconds, it will not happen.
Simple workflows matter more than feature depth.
What’s Next
Immediate focus
Pilot with a small number of NHS training units to see whether real-time broadcasting increases resident doctor participation in procedures.
Next steps
- ARCP evidence pack integration
- Rota-aware opportunity scheduling
- Offline capability for low connectivity areas in hospitals
Longer term
Expand across specialties, provide dashboards for postgraduate medical education teams, and release an Android version.
The goal is simple: make peer learning visible, and capture it the moment it happens.
Built at the Claude AI Hackathon — Track 1: Biology & Physical Health · Track 3: Economic Empowerment & Education
Built With
- core-location
- ios
- openaiapi
- supabase
- swift
- swiftui
Log in or sign up for Devpost to join the conversation.