Senior Full Stack AI Engineer - Voice AI
Description
Build production-grade, real-time voice experiences from streaming audio to user-facing products.
The Opportunity:
We are hiring senior engineers who can own real-time Voice AI applications end to end: Python voice agents, streaming audio pipelines, backend services, cloud infrastructure and modern web interfaces. You will work directly with US-based clients and cross-functional product teams, translating requirements into reliable, low-latency software.
This is a hands-on product engineering role, not a research-only ML role or a chatbot proof-of-concept position.
What You Will Build and Own:
- Develop production voice agents using Python and LiveKit, Pipecat or equivalent real-time frameworks.
- Build streaming speech recognition to LLM / translation to speech synthesis pipelines with low-latency responses.
- Engineer audio ingestion, routing, session lifecycle, multi-party conversations, WebRTC / telephony integration and multilingual workflows.
- Solve turn detection, voice activity detection (VAD), interruptions, buffering, provider failures and recovery.
- Design Python backend services, asynchronous APIs, WebSockets, webhooks, event-driven pipelines and secure data flows.
- Ship responsive user interfaces using React / Next.js and TypeScript / JavaScript.
- Own product features across data models, third-party integrations, deployment and production troubleshooting.
- Integrate, evaluate and optimize LLM, ASR, TTS and translation models with appropriate prompts, fallbacks and performance trade-offs.
- Contribute to AWS-based deployments, Docker, CI/CD, logging, observability, testing and system security; protect sensitive audio and text data.
- Work closely with product stakeholders and US clients in architecture discussions, demos, planning and progress reviews.
Must-Have Qualifications:
- 5+ years of software / full-stack / AI engineering experience, including shipping AI-enabled products to production.
- Demonstrable production Voice AI experience with streaming ASR/TTS, turn-taking, audio transport and low-latency conversational systems.
- Strong Python skills, including asynchronous programming, APIs and production voice agent development.
- Practical full-stack capability: React / Next.js, TypeScript / JavaScript, backend integration and relational databases.
- Experience with LiveKit, Pipecat or a comparable framework; understanding of WebRTC, streaming / event-driven architecture and audio processing.
- Ability to deploy, observe, debug and optimize applications on AWS or comparable cloud using containers and CI/CD.
- Strong spoken and written English is mandatory. Able to independently collaborate with and explain technical decisions to US-based customers.
- Independent problem-solving, sound system design judgment and end-to-end ownership in a fast-moving environment.
Good to Have:
- Go backend experience, or willingness to learn Go; SIP / Twilio / telephony infrastructure.
- ElevenLabs, Deepgram, Whisper, OpenAI Realtime, Vapi, Bland AI, Daily.co or comparable speech platforms.
- LangGraph / LangChain, agentic tool calling, MCP, RAG, model evaluation or inference observability.
- Multilingual speech, translation, speaker diarization, healthcare technology or secure regulated workflows (HIPAA / SOC 2 familiarity).
You Will Thrive Here If:
You have shipped something that people can speak to in real time, understand its failure modes, and can build the product interfaces, APIs and infrastructure around it. You are comfortable owning a complex feature and discussing requirements and trade-offs directly with clients.
Skills
Sign in to read and join the discussion.