Voice
Agent
Production AI for healthcare workflows
Production AI for healthcare workflows
I built and deployed a production Python voice agent for customer-facing healthcare workflows, focusing on multi-intent conversations, tool execution reliability, conversation-state tracking, and deployment isolation. The system reduced escalation errors, lowered multi-intent latency, and made the agent cheaper to operate at call volume.
Role
Software Engineer Intern
System Surface
Voice AI, tool calls, state
Infrastructure
GCP and Kubernetes
Delivery Model
Tenant-isolated deployment
The agent became more dependable through schema redesign, prompt iteration, and conversation-state tracking. Successful AI tool execution moved from 70 percent to 85 percent while cost per 1,000 calls dropped by 27 percent. Escalation errors fell by 60 percent, multi-intent latency came down by 300ms, and tenant-level pod isolation cut deployment time to under 5 minutes.
Study Period
2025 Internship Cycle
Project Status
Rolled out to customers
Primary Focus
Agent reliability
Output Format
Production AI system