Data Machine Learning Engineer
Core
Building end-to-end data pipelines to transform raw studio-quality conversational audio into clean, versioned, training-ready datasets for ultra-realistic voice models.
Role type
Senior Machine Learning Engineer (Data Infrastructure)
Builds
Production data pipelines and training infrastructure for speech AI models
Domain
Speech AI / Audio Technology
Deliverable
production ML models
Required skills
ML data pipeline engineering, speech/audio data processing, speech representations (neural codecs, semantic tokens, mel features), multi-channel audio handling, diarisation, data alignment
Preferred skills
Multilingual data pipelines, large-scale training infrastructure (FSDP, DeepSpeed, Ray), annotation tooling, human-in-the-loop systems
Technologies
FSDP, DeepSpeed, Ray
Responsibilities
Own end-to-end data pipelines from raw audio ingestion to training-ready datasets, build quality systems to catch annotation errors, maintain GPU-fed training infrastructure, iterate on speech representation tooling, handle full- and half-duplex pipeline work including alignment
Seniority
Senior, hands-on IC
