Machine Learning Engineer (Real-Time Speech Translation)

What you'll do

• Develop high-throughput audio and text streaming services, including session, signal-processing, and concurrency management.
• Integrate and serve speech and translation models; improve confidence workflows and real-time updates.
• Scale GPU-backed Kubernetes infrastructure and measure and optimize latency, throughput, and cost.

What they ask for

• BS or MS in Computer Science or a related field, or equivalent practical experience.
• 3+ years building production backend or ML serving systems in Python, including strong asyncio skills.
• Experience with real-time streaming transport (WebSocket or gRPC bidirectional streaming) and production GPU model serving with Docker and Kubernetes.
• Experience integrating speech or NLP models into production; profiled, instrumented, and optimized a real-time or low-latency system.

Nice to have

• Ray Serve, simultaneous or incremental machine translation, or production confidence estimation.
• RabbitMQ or similar, streaming text-to-speech, WebRTC/SFU, CJK text handling, or Datadog/Prometheus.

Pay

$120,000 – $161,434 USD base salary

Tools

Python, asyncio, WebSocket, gRPC, Ray Serve, Triton, vLLM, Kubernetes

AI focus

Build a real-time speech translation system integrating streaming speech recognition and machine translation; use AI coding agents in development. The posting also describes AI-assisted application evaluation.