Principal ML Engineer — Real-Time Voice & Inference

Inworld, Inc.; Mountain View (CA); Presencial; USD 270.000 - 500.000; Jornada completa

Ventajas ofrecidas por este puesto de trabajo

Relocation assistance

Competitive compensation package

Descripción de la vacante

Inworld, Inc. in Mountain View, CA is seeking a Staff/Principal ML Engineer to advance real‑time, scalable voice models.

You will drive inference optimization, model acceleration, and high‑performance backend systems, taking models from research through production with containerization and robust serving. A strong track record in distributed systems, production‑level code, and open source contributions is expected.

Formación

• Inference optimization expertise with scalable serving
• Experience with model acceleration techniques including quantization and distillation
• Proficiency in profiling and optimizing code for NVIDIA GPUs
• Distributed systems experience with Kubernetes, Ray and multi‑node inference
• End‑to‑end ownership from research to production deployment
• PhD or equivalent practical experience in backend or ML systems

Responsabilidades

• Take a model from research to containerized production deployment
• Optimize serving latency and throughput while maintaining reliability
• Own full lifecycle for models in production and scale to thousands of concurrent connections

Conocimientos

Inference optimization

Model acceleration

High‑performance systems

Distributed systems & scaling

Public work

Full‑cycle ownership

Background

Educación

PhD in CS, Physics, Math or equivalent

Herramientas

C++

CUDA

Rust

Python

Kubernetes

Ray

Descripción del empleo

Inworld, Inc. in Mountain View, CA is seeking a Staff/Principal ML Engineer to advance real‑time, scalable voice models.

You will drive inference optimization, model acceleration, and high‑performance backend systems, taking models from research through production with containerization and robust serving. A strong track record in distributed systems, production‑level code, and open source contributions is expected.