Staff Machine Learning Engineer, Voice AI

📋 Description

• Own the voice inference roadmap end-to-end for STT, TTS, and speech-to-speech.
• Drive best-in-class inference performance for voice workloads (latency and throughput).
• Lead productionization of voice models at scale with serverless and dedicated endpoints.
• Build the voice evaluation platform for WER, latency, and naturalness across accents and noise.
• Shape architecture for next-gen model support (audio-native LLMs, SNAC/Encodec).
• Serve as technical DRI for model partner integrations (Cartesia, Deepgram, Rime).

🎯 Requirements

• 8+ years of ML engineering experience focused on model serving at production scale.
• Deep expertise in LLM serving engines (vLLM, SGLang, TensorRT-LLM) or equivalent.
• Expert-level Python and PyTorch; strong GPU optimization (CUDA, profiling).
• Proven system design judgment with scalable architectural decisions.
• Strong technical leadership; high autonomy and engineering quality.
• Familiarity with audio codecs/tokenization (SNAC/Encodec/DAC) is a plus.

🎁 Benefits

• Startup equity
• Health insurance