Speech AI Engineer: TTS & Voice Modeling

Cantina Labs; United States; Remote; USD 180.000 - 260.000; Vollzeit

Benefits dieser Stelle

Medical, dental, and vision insurance

Equity

PTO 42 days incl. PTO, sick days, and

Parental leave & fertility support

401(k) retirement savings plan

Lifestyle spending account – $500/mo

Complimentary lunch and snacks

One Medical membership

Zusammenfassung

Cantina, a new social platform founded by Sean Parker, is seeking a Research / ML Engineer to join our Speech Team. You will build state-of-the-art speech systems end-to-end—from data specs through production inference—and drive the model data evaluation flywheel for TTS and related tasks.

This role blends cutting-edge research with practical engineering, requiring independence to lead projects, design experiments, and deliver reliable, cost-aware models at scale on cloud fleets.

Qualifikationen

• Experience building large-scale audio models (≥3B parameters) and datasets.
• Strong background in transformer and/or diffusion architectures.
• Experience with multi-node and multi-GPU distributed training.
• Proven ability to ship ML systems to production.
• Proficient Python and PyTorch with production-grade code.
• Experience with voice cloning / speech control.
• Publications or notable OSS work in speech/audio/ML.
• Data handling for large ML datasets.

Aufgaben

• Model building for large-scale speech models.
• Lead small research projects.
• Design and analyze scientific experiments.
• Develop dev tooling to improve team productivity.
• Contribute to full-stack optimizations across the stack.
• Define data requirements and data strategies.
• Design automated evaluations and listening tests.
• Harden training → evaluation → inference pipeline.
• Collaborate with infra to scale GPU training/inference.
• Contribute to safety guardrails and responsible speech tech.

Kenntnisse

Large-scale audio models

Transformer architectures

Distributed training

PyTorch

Production-grade code

Voice cloning / speech control

ML data handling

Publications / OSS

Data iteration

Tools

PyTorch

CUDA

C++

Jobbeschreibung

Cantina, a new social platform founded by Sean Parker, is seeking a Research / ML Engineer to join our Speech Team. You will build state-of-the-art speech systems end-to-end—from data specs through production inference—and drive the model data evaluation flywheel for TTS and related tasks.

This role blends cutting-edge research with practical engineering, requiring independence to lead projects, design experiments, and deliver reliable, cost-aware models at scale on cloud fleets.