Senior AI Algorithm Engineer in oneDNN
Core
Design, develop, and optimize low-level performance algorithms and features for oneDNN, a cross-platform open-source deep learning library, targeting Intel CPUs, integrated graphics, and discrete GPUs.
Role type
Senior IC software development engineer (AI algorithms & low-level performance)
Builds
oneDNN library (production ML models)
Domain
AI / Deep Learning / High-Performance Computing / Intel Hardware
Deliverable
production ML models
Required skills
C++, algorithms and data structures, low-level performance optimizations, multithreading, vectorization, memory-layout optimizations, assembly-level programming, Linux software development
Preferred skills
High-performance computing (HPC) applications development, machine learning and deep learning algorithms, software libraries design and architecture, linear algebra solvers, matrix-vector operations, Fast Fourier Transforms, GPU optimizations (OpenCL, CUDA, SYCL/DPC++), parallel programming (OpenMP, TBB, MPI)
Technologies
oneDNN, OpenVINO, TensorFlow, PyTorch, ONNX Runtime, OpenCL, CUDA, SYCL/DPC++, OpenMP, TBB, MPI
Responsibilities
Design and develop new features and algorithms for oneDNN targeting Intel processors and GPUs; Perform performance analysis and optimization for deep-learning inference and training; Develop hardware-specific parallel algorithms including multithreading and vectorization; Contribute to assembly-level programming and low-level performance tuning for Intel microarchitectures; Collaborate with cross-functional teams to ensure integration with Intel's AI ecosystem; Engage with the open-source community and maintain coding standards
Seniority
Senior, hands-on IC
