Machine Learning Engineer, Voice Conversion

What you'll do

• Develop and train large-scale speech models, including pre-training, fine-tuning, and alignment.
• Define data strategies and design objective and subjective evaluations, including listening tests and safety checks.
• Deliver reliable production pipelines by optimizing latency, memory, and cost and supporting monitoring and rollback.

What they ask for

• Exceptional experience with large-scale audio models (>8B parameters, >500k hours of data).
• Hands-on experience with diffusion and/or flow-matching transformers and training audio VAEs, neural audio codecs, and vocoders.
• Experience with multi-node, multi-GPU distributed training and strong PyTorch and software engineering skills.
• Have shipped large-scale speech/audio or multimodal generative models to production.

Nice to have

• Experience with voice cloning, speech control/steerability, or expressive speech generation.
• Notable publications and/or open-source contributions in speech/audio/ML.

Tools

PyTorch, FSDP, DeepSpeed, CUDA, Triton, C++, GRPO, DPO

AI focus

Research, train, evaluate, and deploy safe, steerable speech models for voice conversion and related tasks such as controllable TTS and voice design.