Senior test engineer building AI evaluation systems for automotive voice assistants.
Design and operate an intelligent end-to-end testing platform for AI-powered in-vehicle voice assistants running on Azure. You'll build scoring engines that evaluate non-deterministic LLM outputs using embeddings, LLM-as-judge patterns, and automotive-specific quality metrics (hallucination rate, faithfulness, latency). The role spans prompt regression testing, synthetic dialogue generation with realistic speech variations, and conventional test automation (unit/integration/system tests), plus hands-on validation in vehicles and test environments. Requires deep hands-on expertise in LLM evaluation, Python test frameworks, and ASR/TTS integration; German B2+ is strongly preferred as the project operates primarily in German.
Found at a specialist agency · listed 9 October 2026 · InsideJobs links you to the original posting.