Senior ML Engineer, Data Infrastructure & Pipelines

Unity Technologies; Mountain View (CA); On-site; USD 200,000 - 261,000; Full time

Benefits offered by this job

Health insurance

Stock options

Retirement plan

Vacation days

Parental leave

Commute subsidy

Job summary

Unity Technologies in Mountain View, CA, seeks a Senior Machine Learning Engineer focused on data infrastructure to design and evolve a large-scale offline platform for training data, feature generation, and experimentation workflows.

You will build reliable batch and streaming pipelines using Flink, Spark, and Ray, integrate with Flyte/Airflow, and work closely with ML engineers to scale models and ensure cost-efficient operations.

Qualifications

• Experience with distributed big data processing frameworks (Flink, Spark, Ray).
• Experience building infrastructure for ML training data generation or feature pipelines.
• Experience optimizing pipelines/infrastructure for cost efficiency.

Responsibilities

• Develop infrastructure that supports batch and streaming big data processing (Flink, Spark, Ray).
• Design and operate large-scale data pipelines generating training datasets.
• Integrate data pipelines with workflow orchestration systems (Flyte, Airflow).
• Improve reproducibility and observability of data pipelines via validation/monitoring/testing.
• Optimize performance and resource use across distributed compute systems.
• Collaborate with ML engineers to enable large-scale experimentation and model iteration.
• Lead architectural improvements for scalable, reliable offline pipelines.

Skills

Flink/Spark/Ray

Training data infra

Cost optimization

Python

Data infrastructure

Distributed systems

Technical leadership

Tools

Flyte

Airflow

Job description

Unity Technologies in Mountain View, CA, seeks a Senior Machine Learning Engineer focused on data infrastructure to design and evolve a large-scale offline platform for training data, feature generation, and experimentation workflows.

You will build reliable batch and streaming pipelines using Flink, Spark, and Ray, integrate with Flyte/Airflow, and work closely with ML engineers to scale models and ensure cost-efficient operations.