Founding AI Voice Engineer for Real-Time Mental Health Agents

SaaS Ventures; United States; Remote; USD 120,000 - 180,000; Full time

Job summary

SaaS Ventures is seeking a Founding AI Voice Engineer to build a voice-first AI companion for mental health and personal development. This early-stage effort will operate with no engineering team or PM, moving from prototype to production with rapid iteration and full ownership.

You will own real-time voice-to-voice systems, select LLMs/TTS models, and implement guardrails, safety, and performance optimizations as the product scales.

Qualifications

• Experience building real-time voice-to-voice AI agents.
• Production deployment of such agents in real-world settings.
• Hands-on with STT, LLM and TTS pipelines and/or native speech models.
• Able to handle long, dynamic conversations and ensure reliability.
• Ownership of end-to-end deployment, monitoring and scale.

Responsibilities

• Build and ship iOS & Android apps for the product.
• Develop real-time voice-to-voice AI agents for affective dialogue.
• Select and integrate LLMs and voice models.
• Build RAG, memory and context systems.
• Develop AI personas, behavior and conversational logic.
• Implement safety guidelines and guardrails.
• Translate psychology frameworks into AI behavior with experts.
• Create evaluation, monitoring and testing systems.
• Own backend, infrastructure and deployment.
• Optimize latency, quality, reliability and inference cost.
• Take product from prototype to production at scale.

Skills

Real-time voice AI

Speech-to-speech", "voice-to-voice AI

LLM orchestration

Latency management

Tools

STT pipelines

TTS pipelines

RAG & memory systems

Job description

SaaS Ventures is seeking a Founding AI Voice Engineer to build a voice-first AI companion for mental health and personal development. This early-stage effort will operate with no engineering team or PM, moving from prototype to production with rapid iteration and full ownership.

You will own real-time voice-to-voice systems, select LLMs/TTS models, and implement guardrails, safety, and performance optimizations as the product scales.