Size a voice agent pool by concurrent sessions, not by CPU. At 240 calls in the peak hour with a 4.5 minute average handle time you are offering 18 erlangs of traffic, and Erlang B at 1% blocking needs 28 concurrent slots — three workers at LiveKit's documented 10 jobs per 4-core agent server, four once a rolling deploy is allowed for.
What this covers
- How many workers do I need so calls don't get rejected at peak?
- The worked calculation, from calls per hour to worker count
- Substituting a second set of inputs
- The occupancy gap: why CPU looks calm while calls are refused
- Why one worker holds fewer concurrent calls than you expect
This is a technical summary. The full guide — with the tables and worked examples — is on our site: *How Many Voice Agent Workers Do I Need at Peak*.
Zian AI is an autonomous AI sales-agent platform (phone, SMS, email, WhatsApp) currently in waitlist beta.
Top comments (0)