Machine Learning Performance Engineer

Optiver • New York, NY

9d • Onsite

About The Position

Optiver is seeking a Machine Learning Performance Engineer to join their team, focusing on a pivotal AI initiative. This role offers the opportunity to have a significant impact across Machine Learning infrastructure, training, and inference challenges to advance their futures trading strategies.

Requirements

• Strong knowledge of low-level GPU programming with CUDA, including Tensor Cores, cooperative groups, graphs, and warp-level intrinsics
• Expertise in internals of deep-learning frameworks like PyTorch, JAX, TensorFlow, etc.
• Deep understanding of computer architecture
• Experience in C++ and Python

Nice To Haves

• Experience with JAX ecosystem (XLA, Flax, etc.)
• Familiarity with GPU libraries and tools such as Triton, CUB, cuDNN, and cuBLAS
• Linux system programming experience
• Experience with large-scale distributed training
• Contributions to open-source projects related to data science and machine learning

Responsibilities

• Build scalable and robust training and inference pipelines for deep learning
• Dive into internals of open-source deep learning frameworks and enhance their functionality
• Identify and eliminate performance bottlenecks
• Collaborate closely with researchers and other engineers
• Develop an in-depth understanding of trading systems

Benefits

• Global profit-sharing pool and performance-based bonus structure
• 401(k) match up to 50%
• Comprehensive health, mental, dental, vision, disability, and life coverage
• 25 paid vacation days alongside market holidays
• Extensive office perks, including breakfast, lunch and snacks, regular social events, clubs, sporting leagues and more

What This Job Offers

• Job Type: Full-time
• Career Level: Senior
• Education Level: No Education Listed