WebRTC (Web Real-Time Communication) is an open standard and browser API that enables peer-to-peer audio, video, and data exchange without a plugin, making it the default transport for any live learning feature that needs sub-second latency. The stack has three core components: media capture and encoding (handled by the browser's media engine), ICE (Interactive Connectivity Establishment) for NAT traversal using STUN and TURN servers, and DTLS/SRTP for encrypted media transport. Signaling — the exchange of SDP (Session Description Protocol) offers and answers that describe codec choices and network candidates — is deliberately outside the WebRTC spec, so each platform implements its own signaling channel over WebSocket or HTTP. For live learning the critical property of WebRTC is conversational latency, roughly 150 ms end-to-end as a ballpark target, which is low enough for natural turn-taking between instructor and learner. Screen sharing uses the same WebRTC machinery via the getDisplayMedia API, producing a separate high-resolution track. In a multi-participant class, a direct mesh of peer connections does not scale beyond a handful of participants; an SFU is introduced to route streams efficiently. WebRTC is also the underlying transport for breakout rooms, hand-raising data channels, and in some implementations the whiteboard sync stream.
