I’m building a secure telemedicine application where doctors and patients can meet through high-quality WebRTC video. The back end will be C# (.NET 6 or later), SignalR will handle the real-time signalling layer, and MongoDB will store user, session, and medical data. Core scope • Live video consultations are the heart of the platform; calls must connect quickly and remain stable even on modest bandwidth. • Two-factor authentication is mandatory for every login to protect sensitive health information (SMS or authenticator app, whichever integrates cleanly). • Inside each call participants need screen sharing, a text chat pane, and the ability to record the session (recordings saved to an S3 bucket or GridFS—open to whichever is simpler within MongoDB). Architecture expectations The signalling workflow should stay serverless for media, with STUN/TURN used only when peer-to-peer fails. All real-time messaging (call invites, chat, presence) should flow through SignalR hubs. MongoDB will track user accounts, appointment metadata, and a pointer to each recording. Deliverables 1. Source-controlled solution (Git) containing the C# back end, WebRTC front end (React or Blazor is fine), and a README for local setup. 2. End-to-end demo showing doctor and patient connecting, authenticating via 2FA, starting a video call, sharing a screen, chatting, ending the call, and seeing the recording saved. 3. Brief deployment guide for Azure or AWS (containerised or App Service). 4. Post-handover support for bug fixes during initial testing window. Acceptance criteria • Video latency under 300 ms on a 1.5 Mbps line. • 2FA enforced on every login attempt. • Screen share, chat, and recording all functional during the same session without dropping the call. • All personally identifiable health data encrypted at rest in MongoDB. I value clean, well-commented code and concise commits; please keep third-party libraries to reputable, actively maintained packages only.