Software Engineer-Runtime
Core
Designing, developing, and optimizing core runtime software for AI accelerator hardware to deploy and execute AI models efficiently.
Role type
Systems software engineer (AI accelerator runtime)
Builds
Low-level drivers, firmware, embedded software, and runtime components for AI hardware
Domain
Hardware-software integration, AI infrastructure, embedded systems
Deliverable
production ML models
Required skills
C/C++ programming, systems programming, embedded systems development, computer architecture knowledge, performance profiling, memory management, multithreading, debugging complex hardware-software interactions
Preferred skills
AI accelerator runtime development, AI/ML frameworks (TensorFlow, PyTorch, ONNX Runtime), parallel computing, GPGPU programming (CUDA, OpenCL), embedded Linux/RTOS development, compiler concepts
Technologies
CUDA, OpenCL, TensorFlow, PyTorch, ONNX Runtime, Linux, RTOS
Responsibilities
Design and optimize runtime software components for AI accelerator hardware; Develop and maintain low-level software including drivers and firmware; Optimize runtime performance for latency, throughput, and memory; Collaborate with hardware engineers to define interfaces; Integrate AI models and operators; Debug hardware-software interactions; Write unit and integration tests; Research new technologies for runtime efficiency
