🛠 Tech Stack

💼 About This Role

You'll design and evolve the large-scale offline ML platform at Unity, building reliable infrastructure for training datasets and distributed model training. Your work will directly impact scalable ML workflows used by teams across the company. This role offers the chance to shape the offline ML platform for a leading game engine powering 3 billion monthly consumers.

🎯 What You'll Do

• Design and operate large-scale data pipelines for training datasets
• Develop infrastructure for distributed training workflows using PyTorch, Ray Data, Ray Train
• Integrate ML pipelines with workflow orchestration systems like Flyte or Airflow
• Improve reproducibility and observability through dataset validation and monitoring
• Optimize performance and resource utilization across distributed compute systems

📋 Requirements

• Strong experience building large-scale ML pipelines
• Experience with distributed computing frameworks such as Ray, Spark, Flink and Ray ecosystem
• Deep experience designing and operating production-grade data pipelines
• Strong programming skills in Python and large-scale distributed workloads

✨ Nice to Have

• Experience with modern data infrastructure like data lakes, warehouses, streaming platforms
• Strong systems thinking and ability to reason about performance, scalability, reliability
• Proven ability to lead technical direction and influence architectural decisions

🎁 Benefits & Perks

• 🏖️ Comprehensive health, life, and disability insurance
• 🚗 Commute subsidy
• 📈 Employee stock ownership
• 💰 Competitive retirement/pension plans
• 🌴 Generous vacation and personal days