Loading...
A transparent, real-time view of the Switchboard platform. We publish the health of every customer-facing system, ninety days of uptime history, and a plain-language log of every incident we've resolved — because the businesses and governments that route their front desk through us deserve nothing less.
Each row is a live component and its rolling 90-day history. Hover or focus any single day to see exactly what happened — uptime, state, and the note from that day's incident if there was one.
Speech-to-text, reasoning, and text-to-speech that carry a live conversation.
Inbound and outbound carrier connectivity, call routing, and SMS delivery.
Two-way calendar availability and appointment writes to Google, Calendly, and Acuity.
Retrieval over your business answers, policies, and routing rules.
Operator console — call logs, transcripts, settings, and live monitoring.
Outbound event delivery to your CRM, ticketing, and custom endpoints.
US-only, access-controlled boundary for the FedRAMP-aligned government tenant.
Google Calendar’s API returned 5xx errors for roughly half an hour. Switchboard queued every appointment write and replayed them on recovery — no booking was lost, and callers received confirmations once the upstream cleared.
One carrier region showed elevated packet jitter. Traffic was automatically rerouted to a healthy path; no calls dropped and audio quality stayed within thresholds for the duration.
A downstream customer endpoint timed out and saturated a retry queue. We isolated the slow destination and drained the backlog; all events were delivered with at-least-once guarantees intact.
Our speech-to-text provider reported degraded latency in one region. Conversations stayed live throughout; the pipeline shifted to a secondary lane and latency returned to baseline automatically.
Status pages are easy to make green. We'd rather make ours honest — every bar, every incident, every minute of recovery, in public.