Case Study: SOS.Support -- AI Tech Support for Wholesale Computers
SOS.Support is a live production deployment of goldclaw, handling real customer calls for Wholesale Computers and Technology. This is not a demo -- it's running right now.
The Challenge
Wholesale Computers and Technology needed 24/7 support coverage for both tech support and sales inquiries. Traditional call centers were expensive, and most AI phone solutions required extensive custom development.
The goldclaw Solution
Using goldclaw, SOS.Support built a complete voice support system in three markdown files. The system greets callers, understands whether they need tech support or want to make a purchase, routes them appropriately, and can have natural free-form conversations powered by an LLM.
The Architecture
PROCEDURES.md defines the voice capabilities: Say (TTS with 24kHz to 16kHz resampling), Listen (STT), Analyze (intent classification), and Execute (shell commands for CRM integration). WORKFLOW.md defines the call flow as a state machine. MODULES.md wires VoiceHandler + ToolHandler into a SuperAgent.
$ goldclaw status
Workflow Engine compiled successfully
VoiceHandler module compiled from WORKFLOW.md
Resilience Plugin registered
AMCP connected to RabbitMQ
TTS is working and resampling from 24kHz to 16kHz
Common phrases are being pre-cached in the background
Uptime: 47 days, 3 hours
Calls handled: 12,847
Avg response time: 340msResults
24/7 coverage with no human operators needed for routine inquiries. Average response time under 400ms. Self-healing architecture means zero manual intervention for infrastructure issues. The entire system is maintained by editing three markdown files.