Staff Software Engineer, Feature Platform & ML Infra

Unity3d; Bellevue (WA); On-site; USD 192,600 - 288,800; Full time

Job summary

Unity is hiring a Staff Software Engineer for the Feature Platform to own end-to-end data pipelines powering ads ML workflows. You’ll design, build, and operate scalable systems transforming high-volume events into production datasets for bidding, attribution, and ranking.

You’ll collaborate with ML, backend, and product teams to deliver durable infrastructure, from offline training to online feature serving, while emphasizing correctness, observability, and performance at scale.

Qualifications

• Strong fundamentals in large-scale distributed systems.
• Hands-on ETL/ELT pipelines in Python/Java/SQL.
• Experience with Spark or Flink in batch and streaming.
• Understanding offline data systems and online serving layers.
• Cloud-native environments, containerization and Kubernetes.

Responsibilities

• Design, build, and operate scalable, production-grade data pipeline systems and curated feature datasets powering ads optimization and ML.
• Own end-to-end offline data flows from raw event ingestion to feature-ready datasets with emphasis on correctness and SLA compliance.
• Develop and maintain large-scale batch and streaming systems (Python/Java/SQL) with focus on performance and reliability.
• Build and contribute to the Feature Store platform and online serving layer.
• Drive observability, testing frameworks, and quality standards across the platform.
• Lead architectural decisions and engineering best practices within the team.

Skills

Distributed systems design

ETL/ELT pipelines

Python

Java

SQL

Go

Kubernetes

Spark

Flink

Tools

Kubernetes

Go-based services

Job description

Unity is hiring a Staff Software Engineer for the Feature Platform to own end-to-end data pipelines powering ads ML workflows. You’ll design, build, and operate scalable systems transforming high-volume events into production datasets for bidding, attribution, and ranking.

You’ll collaborate with ML, backend, and product teams to deliver durable infrastructure, from offline training to online feature serving, while emphasizing correctness, observability, and performance at scale.