Run voice agents in production without flying blind.
Small tools for the problems teams hit after the demo works: what each call costs, how long callers wait, why SIP calls fail, what broke since the last prompt change. Pick the one that hurts most.
The tools
Each one solves one problem and works on its own. VoiceOps puts cost, latency and SIP in one dashboard.
VoiceOps
One collector, one dashboard: what each call cost, how long the caller waited, why SIP failed.
Open Open sourceTally
Cost and margin per voice-agent call and per customer, across STT, LLM, TTS and phone bills.
Open Open sourceHold
A turn-by-turn timeline of the wait your callers hear, and the step where it is lost.
Open Open sourceLine
Watches your numbers and SIP trunks and tells you why calls will fail, before they do.
Open Open sourceRing
Calls your agent over the real phone network every few minutes and alerts when it is slow or silent.
Open Open sourceRedial
Every bad call becomes a regression test that runs before each prompt or model change.
Open Early accessClearance
Checks consent, do-not-call lists, calling hours and AI disclosure before an outbound call.
Open Calculator · open sourcePortout
How much you would save leaving a hosted voice platform, and a port of your agent to LiveKit.
Open ServiceTeardown
We review your voice agent: what it overspends on, where it loses latency, what breaks in telephony.
OpenWhat hurts most right now?
Pick one. We build what most teams ask for first, and we will write to you when it is ready for your calls.
We only use your email to contact you about this. Privacy.