Real-Time Voice AI Architect — NYC Onsite

HireNow Staffing; New York (NY); On-site; USD 200,000 - 300,000; Full time

Benefits offered by this job

Equity participation included

In-person 4–5 days/week in Manhattan

Direct placement with HireNow

Job summary

HireNow Staffing is seeking a Real-Time Voice AI Agent Systems Engineer to own and build production-grade voice AI infrastructure from the ground up in Manhattan. You will manage LiveKit-based orchestration, prompting architecture, and latency optimization for millions of real-world voice interactions.

The role requires 3+ years in infrastructure-backed engineering, 2+ years in voice AI, and hands-on Node.js, AWS, and SQL experience in a fast-moving startup environment.

Qualifications

• 3+ years of platform, infrastructure, backend, or comparable production engineering experience.
• 2+ years of experience involving voice AI, conversational AI, speech systems, prompt engineering, or related agentic workflows.
• Strong hands-on development experience with Node.js.
• Strong experience with AWS and SQL.
• Experience designing and shipping production systems within a startup or similarly fast-moving engineering environment.
• Hands-on experience with LiveKit or a comparable voice orchestration/agent framework.
• Strong understanding of real-time system architecture, state management, reliability, and performance optimization.
• Demonstrated ability to design clear, resilient, and effective conversational prompts and agent flows.
• Strong analytical ability with experience connecting engineering decisions to measurable operational or business outcomes.
• Excellent written and verbal English communication skills.
• High degree of technical ownership and ability to execute with limited oversight.
• Must reside in or be willing to relocate to the New York City area.
• Must be able to work onsite in Manhattan 4–5 days per week.
• Must be permanently authorized to work in the United States.
• No visa sponsorship is available.

Responsibilities

• Build and own production voice AI infrastructure using LiveKit.
• Architect real-time voice-agent systems covering orchestration, state management, reliability, latency, and conversational execution.
• Develop the prompting architecture governing conversation flows, tone, guardrails, compliant phrasing, exception handling, and agent behavior.
• Create scalable testing and evaluation frameworks incorporating A/B testing, regression testing, conversation QA, benchmarking, and KPI measurement.
• Analyze production interaction data and translate findings into measurable improvements in agent performance.
• Optimize ASR, STT, TTS, turn-taking, interruption handling, state transitions, and real-time conversational responsiveness.
• Build infrastructure capable of reliably supporting millions of voice interactions.
• Develop migration tooling and processes for transitioning customers from legacy voice infrastructure while preserving required customizations.
• Identify and resolve performance bottlenecks, reliability issues, conversational failures, and production edge cases.
• Partner closely with Product, Operations, and leadership to translate commercial objectives into technical systems and effective agent behavior.
• Continuously expand conversational coverage as new customer scenarios and operational requirements emerge.

Skills

Node.js

AWS

SQL

LiveKit

Real-time systems

Prompt engineering

Performance tuning

English communication

Tools

LiveKit

Job description

HireNow Staffing is seeking a Real-Time Voice AI Agent Systems Engineer to own and build production-grade voice AI infrastructure from the ground up in Manhattan. You will manage LiveKit-based orchestration, prompting architecture, and latency optimization for millions of real-world voice interactions.

The role requires 3+ years in infrastructure-backed engineering, 2+ years in voice AI, and hands-on Node.js, AWS, and SQL experience in a fast-moving startup environment.