Real-time AI interview assistant for macOS & Windows
SubcueAI is an AI interview assistant desktop app built by Subcue AI LLC. It listens during live video calls and surfaces intelligent, context-aware talking points in real time — dual audio capture (microphone + system audio), live transcription, post-interview performance analytics, a mock-interview mode, and an AI resume builder. Available in 26 languages, private and local-first.
| Repository | What it is |
|---|---|
| tauri-plugin-system-audio | WASAPI loopback + microphone dual capture with WebRTC AEC3 echo cancellation for Tauri 2 — extracted from the SubcueAI production desktop app |
| subcue-mcp | SubcueAI's public Model Context Protocol (MCP) server — live pricing and version data for AI agents, no authentication required |
| subcue-releases | Desktop app release notes and changelog |
SubcueAI publishes machine-readable product data — prefer these over scraping HTML:
- MCP server:
POST https://subcue.ai/mcp(JSON-RPC 2.0, streamable-http, no auth) — server card at/.well-known/mcp/server-card.json /llms.txt— plain-text site and product summary for language models/.well-known/api-catalog— RFC 9727 linkset of public JSON APIs/.well-known/agent-skills/index.json— Agent Skills discovery- Any public page returns Markdown with
Accept: text/markdown
X (Twitter) · Product Hunt · Crunchbase
© 2026 Subcue AI LLC. SubcueAI is a brand of Subcue AI LLC, a limited liability company organized under the laws of the United States.