> Markdown version of [/jobs/ext/2789311-senior-machine-learning-engineer-voice-agents](https://www.wearedevelopers.com/jobs/ext/2789311-senior-machine-learning-engineer-voice-agents). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md). --- # Senior Machine Learning Engineer, Voice Agents - **Company:** Jobgether - **Location:** Germany (Remote available) - **Experience:** Expert - **Contract:** Permanent contract - **Skills:** Application Programming Interfaces (APIs), Artificial Intelligence, User Authentication, Programming Tools, Distributed Systems, Fault Tolerance, Python (Programming Language), Load Testing, Machine Learning, Open Source Technology, Data Streaming, Management of Software Versions, WebSocket, WebRTC, Graphics Processing Unit (GPU), Large Language Models, Concurrency, Backend, Build Management, Live Streaming - **Published:** September 8, 2026 - **Apply:** https://www.adzuna.de/details/5873204226 ## About the Role * Senior-level engineering experience with the ability to independently own a substantial part of an architecture and drive it forward. * Experience building developer-facing infrastructure in AI, machine learning, developer tools, or a comparable technical environment, such as inference APIs or agent infrastructure. * Significant open-source contributions to a Python library and strong proficiency with asynchronous Python. * Solid understanding of distributed systems and their failure modes. * Proven experience shipping realtime technology involving streaming, WebSockets, WebRTC, audio or video pipelines, or live inference. * Practical production experience with LLMs or multimodal models. * Strong written communication skills and a demonstrated ability to collaborate asynchronously and in public. * Genuine interest in voice technology and conversational AI. * Contributions to voice-agent frameworks such as speech-to-speech, Pipecat, LiveKit Agents, Vocode, or TEN are a plus. * Experience contributing to llama.cpp or another low-level inference runtime is advantageous. * Hands-on experience with ASR, TTS, or end-to-end speech models, including evaluating latency and quality trade-offs, is a plus. * GPU serving, quantization, or on-device inference experience is advantageous. * Knowledge of audio pipelines, including VAD, echo cancellation, jitter buffers, barge-in, and turn detection, is a plus. * Experience deploying technology to embedded or robotics environments is beneficial. * A public technical track record through talks, blog posts, demos, or similar contributions is valued. * Candidates are encouraged to apply even if they do not meet every listed requirement, particularly where their experience could bring complementary strengths to the team. ## Description Own a major part of an open voice-agent stack at the intersection of machine learning, realtime systems, and developer infrastructure. You will lead the architecture and evolution of an open-source speech-to-speech library while helping turn a new voice platform into a production-ready developer product. The role combines deep backend and ML engineering with realtime audio, inference infrastructure, and developer experience. You will integrate rapidly evolving ASR, TTS, and end-to-end speech models while maintaining clean abstractions and strong reliability. You will have significant autonomy to shape APIs, streaming protocols, GPU serving, observability, and production architecture. Your work will directly enable developers to build and deploy sophisticated voice agents and will support existing realtime deployments, including robotics applications. This is an open, highly collaborative environment where you can contribute publicly through documentation, demos, talks, and open-source development. Accountabilities * Take architectural ownership of significant parts of the open-source speech-to-speech library, including pipeline design, latency budgets, and realtime loop reliability. * Integrate new ASR, TTS, and end-to-end speech models as they become available while maintaining clean, extensible abstractions. * Review community contributions, triage issues, manage releases, and help grow the contributor community around the project. * Design the developer API and streaming protocol for the voice platform, including session lifecycle, WebSockets/WebRTC transport, authentication, error semantics, and versioning. * Build and operate the serving infrastructure for realtime GPU inference, including concurrency, autoscaling, observability, and cost-per-session optimization. * Collaborate with Hub and inference teams to make voice agents easy to integrate into products, applications, and demonstrations. * Take the platform from prototype to production through load testing, SLO definition, reliability improvements, and graceful degradation when models or network paths fail. * Create documentation, examples, and templates that enable developers to move from initial setup to a running voice agent quickly. * Support deployments already relying on the technology, including the existing robotics fleet. * Contribute to the wider technical community through blog posts, demonstrations, conference talks, or other public technical content when desired. ## Related Videos - [Minimal infrastructure for Real‑Time Phone Agents: transcripts in, responses out](https://www.wearedevelopers.com/videos/1736-minimal-infrastructure-for-real-time-phone-agents-transcripts-in-responses-out) - [On a Secret Mission: Developing AI Agents](https://www.wearedevelopers.com/videos/1510-on-a-secret-mission-developing-ai-agents) - [Developing the Backend with Stefan Lingler, CTO at Shpock](https://www.wearedevelopers.com/videos/100360-developing-the-backend-with-stefan-lingler-cto-at-shpock) - [Transforming Education: A Journey from interactive Markdown to Remote-Labs](https://www.wearedevelopers.com/videos/941-transforming-education-a-journey-from-interactive-markdown-to-remote-labs) - [Raise your voice!](https://www.wearedevelopers.com/videos/10-raise-your-voice) - [Hello JARVIS - Building Voice Interfaces for Your LLMS](https://www.wearedevelopers.com/videos/1641-hello-jarvis-building-voice-interfaces-for-your-llms) ## Related Articles - [What Are Large Language Models?](https://www.wearedevelopers.com/magazine/304-what-are-large-language-models) - [Dev Digest 121 - AI goes offline](https://www.wearedevelopers.com/magazine/456-dev-digest-121-ai-goes-offline) - [How to Become an AI Engineer](https://www.wearedevelopers.com/magazine/331-how-to-become-an-ai-engineer) - [What is Agentic Programming and Why Should Developers Care?](https://www.wearedevelopers.com/magazine/625-what-is-agentic-programming-and-why-should-developers-care) - [A 5-Step Open-Source Setup for Agentic Engineering](https://www.wearedevelopers.com/magazine/738-a-5-step-open-source-setup-for-agentic-engineering) - [Never delegate the understanding](https://www.wearedevelopers.com/magazine/749-never-delegate-the-understanding)