Senior AI Voice Systems Engineer (Orchestrator / Real-Time Systems)

Role Overview

We are seeking a Senior AI Voice Systems Engineer to design, build, and operate a real-time, AI-powered voice orchestration platform.

This role sits at the intersection of:

AI/LLM orchestration
telephony systems
backend engineering
workflow automation
reliability engineering (SRE)

You will be responsible for building and maintaining systems that manage live customer calls, perform real-time decisioning, and integrate AI agents with external services and human support teams.

Key Responsibilities

🧠 AI Orchestration & Conversation Systems

Design and implement multi-step conversational flows using AI voice agents (e.g., Retell or similar platforms)
Structure and optimize system prompts and knowledge bases for performance and maintainability
Implement interrupt handling (e.g., “talk to human”) and dynamic routing during conversations
Ensure AI agents operate reliably under real-world conditions (latency, ambiguity, user interruptions)

⚙️ Backend & Orchestration Logic

Build and maintain orchestration services to manage call state, routing, and decision logic
Implement state management for real-time sessions (Redis, DB, or similar)
Design APIs for: authentication & verification routing decisions workflow triggers
Ensure clear separation between: AI layer (conversation) orchestration layer (decision-making) execution layer (workflows, integrations)

☎️ Telephony & Real-Time Systems

Integrate with telephony platforms (e.g., Jambonz, Twilio, SIP systems)
Handle: inbound/outbound call flows warm and cold transfers failure scenarios and retries
Optimize call latency and responsiveness

🔄 Workflow Automation & Integrations

Design and maintain n8n workflows (or similar tools) for: authentication fallback troubleshooting flows third-party integrations
Integrate with external services (APIs, CRMs, internal systems)
Ensure workflows are idempotent, traceable, and debuggable

⚡ Performance & Reliability (SRE Mindset)

Identify and resolve system bottlenecks (e.g., verification latency, API delays)
Implement fallback strategies: Lambda → workflow fallback warm → cold transfer fallback
Define retry strategies and escalation paths
Ensure high availability and graceful degradation

🧪 Testing & Quality Assurance

Perform end-to-end (E2E) testing using live telephony flows
Validate: AI behavior workflow execution transfer success/failure handling
Work within production constraints (no dedicated QA environment)
Collaborate with QA and business stakeholders to define test scenarios

📊 Observability & Debugging

Implement logging and monitoring across: AI interactions API calls workflows telephony events
Track key metrics: call success rate verification time retry rates transfer failures
Debug cross-system issues spanning multiple services

Required Skills & Experience

Core Engineering

5+ years in backend or distributed systems engineering
Strong experience with: Node.js, Python, or similar REST APIs and event-driven systems
Experience designing stateful systems

AI / LLM Systems

Experience working with: LLM APIs (OpenAI, Gemini, etc.) prompt engineering and optimization
Understanding of: system prompt vs knowledge base design latency and token constraints
Experience building agentic workflows (not just chatbots)

Real-Time / Telephony (Highly Preferred)