Machine Learning Performance Engineer - Offboard Training & Inference

📋 Description

• Profile and optimize distributed training end to end
• Optimize offline and batch inference over petabyte logs
• Establish roofline and performance models
• Improve multi-node scaling efficiency
• Drive cluster goodput and reduce GPU idle time
• Build benchmarking, observability, and regression tooling

🎯 Requirements

• Hands-on ML performance engineering experience
• Experience with distributed multi-node training at scale
• Deep familiarity with GPU performance concepts
• Experience with high-throughput or batch inference systems
• Fluency in Python and proficiency in C++
• Strong debugging, analytical, and problem-solving skills

🎁 Benefits

• Collaborative culture and ownership mindset
• Equal opportunity employer commitments