Voice AI Engineer

<h2>Intro</h2><p><em>For this position we are only looking for <strong>candidates located in Greece, Germany or Austria</strong>.</em></p><p>Nairon GmbH is a Swiss startup building AI voice agents that answer and make phone calls for businesses. Our platform runs real-time conversations over the phone and on the web, connected to our customers' telephony and business tools. We value clear communication, thoughtful engineering, and a collaborative approach to problem-solving. Join us as we continue to develop tools that make a difference.</p><h2>Tasks</h2><p>As a Voice AI Engineer at Nairon GmbH, you will make our agents sound fast, natural and reliable in live calls. Your responsibilities will include:</p><ul><li>Improving turn-taking: when the agent speaks, waits, yields or gets interrupted, including voice activity and end-of-turn detection.</li><li>Cutting end-to-end latency across speech recognition, language models and speech synthesis, and making provider failover seamless.</li><li>Optimising how we use LLMs in conversation: prompting, tool calls, streaming and speculative or parallel requests.</li><li>Serving our own speech models (STT and TTS) on GPUs, measuring throughput and time to first audio.</li><li>Fine-tuning speech and language models for our customers' languages and voices.</li><li>Building evaluation: test calls, benchmarks and quality metrics that tell us when a change really helps.</li></ul><h2>Requirements</h2><p><strong>Experience:</strong></p><p>A senior or strong mid-level engineer with production experience in real-time audio, conversational AI or ML model serving.</p><p><strong>Technical Skills:</strong></p><p>Required:</p><ul><li>Strong <em>Python</em>, including asyncio and streaming pipelines.</li><li>Hands-on work with speech (STT, TTS, VAD) and LLM APIs in latency-sensitive systems.</li><li>Model serving on GPUs, for example with <em>vLLM</em> or similar, and the basics of batching and memory limits.</li></ul><p>Nice to have:</p><ul><li><em>LiveKit Agents</em>, <em>pipecat</em> or WebRTC.</li><li>Fine-tuning (LoRA or full) of speech or language models.</li><li>Audio signal processing such as echo cancellation or noise suppression.</li><li>German, or experience with German-language speech data.</li></ul><p><strong>Communication:</strong></p><ul><li><em><strong>English proficiency is essential</strong>, both written and spoken, as it is the working language for all positions.</em></li></ul><p><strong>Soft Skills:</strong></p><ul><li>You measure before and after, and trust numbers over impressions.</li><li>Ownership of a problem from the first bad call to the fix in production.</li></ul><h2>Benefits</h2><ul><li><strong>Flexible Work Setup:</strong> Work fully remotely or from a co-working space — whichever suits you best.</li><li><strong>Equipment Provided:</strong> Company laptop and phone provided for all employees.</li><li><strong>Collaborative Environment:</strong> Join a small, agile team where your contributions have a direct impact.</li><li><strong>Growth Opportunities:</strong> Work across the whole voice stack, from model serving to the conversation the caller hears.</li></ul><h2>Closing</h2><p>If you want to make AI conversations feel human on a real phone line, we'd love to hear from you.</p><p>Find <a href="https://www.arbeitnow.com">Jobs in Germany</a> on Arbeitnow</a>