Senior ML Engineer, Speech & TTS - Remote

Bazeta; United States; Remote; USD 180,000 - 260,000; Full time

Benefits offered by this job

Health insurance

401(k) retirement plan

20+ days PTO / year

Job summary

Cantina Labs is seeking a senior ML Engineer to join the Speech Team, focusing on building state-of-the-art speech systems from data specs to production inference. You will lead small research projects, design rigorous experiments, and contribute to end-to-end model pipelines for TTS, voice cloning, and related tasks.

You should have experience with large-scale audio models, transformer/diffusion architectures, and multi-node distributed training, with a track record of shipping models to

Qualifications

• Experience with large-scale audio modeling (3B+ params, 500k+ hours data).
• Strong background in transformer and/or diffusion models for speech.
• Experience with multi-node/multi-GPU distributed training.
• Shipped speech/audio models to production and Python/PyTorch proficiency.

Responsibilities

• Architect and train large-scale speech models end-to-end.
• Lead small research projects within the team.
• Design experiments to improve model quality and robustness.
• Develop dev tooling to boost team productivity and maintainability.
• Collaborate across data, research, and infra for scalable production.

Skills

Large-scale audio models

Transformer architectures

Diffusion models

Voice cloning

Distributed training

PyTorch

Performance optimization

Tools

CUDA

Triton

C++

Job description

Cantina Labs is seeking a senior ML Engineer to join the Speech Team, focusing on building state-of-the-art speech systems from data specs to production inference. You will lead small research projects, design rigorous experiments, and contribute to end-to-end model pipelines for TTS, voice cloning, and related tasks.

You should have experience with large-scale audio models, transformer/diffusion architectures, and multi-node distributed training, with a track record of shipping models to