Sr. Model Engineer, Voice
Core
Design, build, and maintain distributed data pipelines and training infrastructure for speech-language models, integrating them into production applications and real-time services.
Role type
Senior IC model engineer (voice/audio)
Builds
Distributed data pipelines, training infrastructure, and production voice models
Domain
Voice technology, large-scale machine learning
Deliverable
production ML models
Required skills
Python, PyTorch, JAX, Rust, C++, distributed systems, large-scale voice products, automated evaluation systems, model optimization
Preferred skills
Kubernetes, JAX/PyTorch optimization, automated testing and A/B experimentation pipelines, distributed data processing with Spark and Ray
Technologies
JAX, PyTorch, Kubernetes, Spark, Ray
Responsibilities
Design and build distributed data pipelines for audio ingestion and annotation; Build and optimize distributed training infrastructure for speech-language models; Develop automated evaluation, testing, and experimentation systems; Integrate voice models into production applications and real-time services; Profile and optimize training and inference performance across throughput, latency, and compute efficiency.
Seniority
Senior, hands-on IC
