Senior ML Infrastructure Engineer – Online Inference

Unity Technologies; United States; Remote; USD 165,000 - 210,000; Full time

Benefits offered by this job

Health benefits

Stock options

Vacation & personal days

Commuter subsidy

Job summary

Unity Technologies is seeking a Senior ML engineer to design and evolve Unity Vector’s online model inference platform. You will build reliable infrastructure for serving ML models in production, optimize inference performance, and enable safe experimentation across high-traffic systems.

You will collaborate with ML engineers, platform teams, and product stakeholders to ensure models are deployed, scaled, monitored, and iterated efficiently, shaping packaging, serving, validation, and deployment

Qualifications

• Production ML inference experience in scalable systems.
• Experience with model serving frameworks and deployment lifecycles.
• Strong software engineering skills in Python and distributed systems.

Responsibilities

• Design and operate large-scale online inference infrastructure with low latency.
• Develop distributed training workflows and integration with orchestration tools.
• Improve observability, reliability, and cost-efficiency of ML serving platforms.
• Lead architectural improvements for a robust, scalable online ML platform.

Skills

Python

Distributed systems

Kubernetes

PyTorch

Model serving

Performance tuning

Canary testing

Observability

Education

Bachelor's degree in CS

Tools

NVIDIA Triton

TorchServe

TensorFlow Serving

Flyte/Airflow

Job description

Unity Technologies is seeking a Senior ML engineer to design and evolve Unity Vector’s online model inference platform. You will build reliable infrastructure for serving ML models in production, optimize inference performance, and enable safe experimentation across high-traffic systems.

You will collaborate with ML engineers, platform teams, and product stakeholders to ensure models are deployed, scaled, monitored, and iterated efficiently, shaping packaging, serving, validation, and deployment