voice AI Engineer

Responsibilities:

• Own the real-time interpretation pipeline end-to-end, including turn detection, streaming speech recognition, context-aware translation, and streaming speech synthesis, delivering subsecond performance in healthcare environments.
• Develop and extend our Python voice agents, including audio routing and session lifecycle management for multi-party, multilingual conversations.
• Integrate and deploy AI/ML models in collaboration with the ML team, from third-party speech and language providers to our own fine-tuned models.
• Build intuitive, responsive UIs using React / Next.js and modern web frameworks.
• Design and optimise streaming workflows, including audio ingestion, parsing, and processing for subsecond performance.
• Build and maintain APIs, webhooks, event pipelines, and data models that drive real-time communication and interaction.
• Develop backend services and APIs in Go supporting session management, configuration, and secure data flows for the voice pipeline.
• Implement best-in-class security and compliance practices for sensitive medical audio and text data.
• Contribute to DevOps and platform engineering, including CI/CD automation, monitoring, logging, and system reliability.
• Collaborate closely with product, design partners, and clinical advisors to ship safe, reliable, production-grade tools for healthcare environments.

Requirements:

• 5+ years of experience as a software, full-stack, or machine learning engineer, including experience shipping AI-powered features to production.
• Strong experience with real-time voice or conversational AI systems, including streaming speech recognition and synthesis, voice activity and turn detection, LLM orchestration, and low-latency processing.
• Strong proficiency in Python, including asynchronous programming, and experience building production voice agents with LiveKit, Pipecat, or a comparable real-time framework.
• Proficiency in Go, or strong experience in another backend language and a willingness to work in Go.
• Experience with React, Node.js, and modern API/web application architecture.
• Hands-on experience with API design, audio processing, event-driven and streaming architectures, and WebRTC or telephony media.
• Experience integrating and evaluating LLMs and speech models in production, including prompt design, provider selection, fallbacks, evaluation, and cost and latency trade-offs.
• Proficiency with AWS, containerised services, and infrastructure-as-code.
• Strong grasp of security, privacy, and the handling of sensitive data in production systems.
• Startup-ready mindset: proactive, fast-moving, comfortable with ambiguity, excited to own problems end-to-end.

Nice to Have:

• Healthcare experience or familiarity with clinical workflows, EHR systems, or regulated environments.
• Experience with voice AI, ML ops, or building model-driven real-time products.
• Background in NLP, speech recognition, speech synthesis, or audio signal processing.
• Deep knowledge of cloud security, privacy, compliance, or enterprise-grade architectures (HIPAA, SOC 2, etc. ).
• Experience scaling data-intensive pipelines or working with multilingual speech datasets.
• Speak a language other than English.

Skills

AWS, Conversational AI, Golang, LiveKit, Next.js, Pipecat, Python, React, WebRTC, voice AI, voice Agent*