Staff Platform Engineer - Data & ML

Unity South APAC (SEA, ANZ, IND Subcont.); Mountain View (CA); On-site; USD 245,000 - 318,000; Full time

Benefits offered by this job

Comprehensive health insurance

Commute subsidy

Employee stock ownership

Generous vacation and personal days

Training and development programs

Job summary

Unity is seeking a Staff Software Engineer for the Feature Platform to design and operate production-grade data pipelines powering ads optimization and ML across a global scale.

You will own end-to-end offline data flows, build batch and streaming systems, and contribute to a feature store with Go-based online serving. Strong emphasis on correctness, reproducibility, and reliability across teams.

Qualifications

• Experience building production-grade ETL/ELT pipelines using Python, Java, SQL.
• Deep experience designing and operating large-scale distributed systems in production.
• Familiarity with Spark or Flink for batch and streaming data processing.
• Knowledge of feature stores, ML infrastructure, and real-time APIs is a plus.

Responsibilities

• Design, build, and operate scalable data pipelines and feature datasets for ads optimization and ML.
• Own end-to-end offline data flows from ingestion to feature-ready data with accuracy and SLA adherence.
• Develop and maintain large-scale batch and streaming systems with a focus on performance and reliability.
• Contribute to the Feature Store platform and integrate with high-throughput online serving layers in Go.
• Translate product and monetization logic into extensible analytics and ML infrastructure.

Skills

Python

Java

SQL

Spark

Flink

Go

Kubernetes

ML infrastructure

Distributed systems

Data pipelines

Tools

Cloud environments

ETL/ELT pipelines

Job description

Unity is seeking a Staff Software Engineer for the Feature Platform to design and operate production-grade data pipelines powering ads optimization and ML across a global scale.

You will own end-to-end offline data flows, build batch and streaming systems, and contribute to a feature store with Go-based online serving. Strong emphasis on correctness, reproducibility, and reliability across teams.