Switchboard
Real-time voice-AI orchestration engine
Routes phone calls (PSTN/SIP) through STT → LLM → TTS with streaming and barge-in. Multi-tenant, sub-300ms turn latency.
- TypeScript
- Node.js
- WebSockets
- Redis
Real-time backend systems and the ML that ships them — plus the tooling around both. Filter by discipline.
Real-time voice-AI orchestration engine
Routes phone calls (PSTN/SIP) through STT → LLM → TTS with streaming and barge-in. Multi-tenant, sub-300ms turn latency.
Transcript intelligence service
Real-time intent, sentiment & entity extraction over call transcripts. TensorFlow → ONNX with dynamic request batching.
Integration middleware framework
A tenant / adapter / router framework to wire CRMs, telephony and webhooks together without per-client spaghetti. Dual deploy (SaaS + self-hosted).
Live agent observability dashboard
Real-time view of running agents: live transcripts, latency graphs, error feed and per-tenant metrics.
LLM / agent evaluation harness
Regression-tests prompts & agent flows like code: fixtures, assertions, LLM-as-judge + deterministic checks, CI gate, HTML reports.
Webhook capture & replay CLI
Capture and replay webhooks for local dev. Small, sharp, open-source — signals CLI/systems maturity and an OSS presence.