Optiver is seeking a Machine Learning Performance Engineer to advance their futures trading strategies by addressing Machine Learning infrastructure, training, and inference challenges.
Requirements
- Expertise in internals of deep-learning frameworks like PyTorch, JAX, TensorFlow, etc.
- Deep understanding of computer architecture
- Experience in programming in C++ and Python
- Experience with JAX ecosystem (XLA, Flax, etc.)
- Experience in programming for GPUs or other accelerators (CUDA, Triton, Pallas, etc.)
- Linux system programming experience
- Experience with large-scale distributed training
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
- Develop an in-depth understanding of trading systems
- Collaborate closely with researchers and other engineers
Other
- Work alongside best-in-class professionals from over 40 different countries
- 25 paid vacation days and market holidays
- Fully paid health insurance
- Daily breakfast and lunch
- Training opportunities
- 401(k) match up to 50% and charitable match opportunities
- Regular social events and clubs
