High Performance Low Level Programmer

What you'll do

• Design and implement micro-architecturally optimized components for the Accelerate framework.
• Develop high-performance, energy-efficient routines for machine learning and vision algorithms.
• Write low-level code using vector assembly and C to optimize CPU core performance.
• Implement computational support for 2D multi-layered convolution, LSTM, FFT, and DFT.
• Optimize software and hardware integration to improve low-level computational efficiency.
• Develop robust algorithms for cryptographic standards such as AES and SHA.
• Support the development of core operating system subsystems for iOS, macOS, watchOS, and tvOS.

What we're looking for

• 5+ years of low level algorithmic development experience in Machine Learning computation, LSTM, BLAS, and DFT.
• Detailed knowledge of vector Instruction Set Architectures (ISA) for both ARM and Intel.
• Excellent coding skills in Assembly (ASM) and C.
• Strong understanding of computational efficiency and micro-architecture.
• Understanding and working knowledge of cryptographic algorithms such as AES and SHA (preferred).
• Ability to craft the fastest and most energy-efficient routines for a particular CPU core (preferred).
• Strong verbal and written communication skills (preferred).
• Ability to manage multiple tasks and self-prioritize (preferred).