7 месяцев назад
NPU Software Engineer - Kernel (AI)
Вакансия из Hirify Global, списка международных tech-компанийДля мэтча и отклика нужен Plus
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Описание вакансии
Текст:
TL;DR
NPU Software Engineer - Kernel (AI): Designing and developing compute libraries and architecture-aware custom kernels for Rebellions’ proprietary NPU instruction set architecture with an accent on neural network operations, heterogeneous compute optimization, and generative AI workloads. Focus on expanding operation coverage, improving numerical stability and performance, and contributing to hardware–software co-design for next-generation compute architectures.
Location: On-site in Seongnam, South Korea, at ’ R-TOWER office.
Company
develops proprietary neural processing units and the software and hardware architectures that support AI workloads.
What you will do
- Design and develop BLAS, DNN, and other compute libraries optimized for ’ proprietary instruction set architecture.
- Expand neural network operation coverage across tensor shapes, precision requirements, and numerical stability constraints.
- Optimize operation-level performance and utilization of heterogeneous compute resources based on workload characteristics.
- Design and implement architecture-aware custom kernels for dynamic generative AI execution patterns.
- Support hardware–software co-design by identifying critical features for next-generation compute architectures.
Requirements
- Master’s degree or higher in Electrical Engineering, Computer Science, or a related field.
- In-depth knowledge of computer architecture and parallel programming.
- Strong understanding of neural network operations, including high-level concepts and low-level computational dataflows.
- Strong analytical, troubleshooting, and performance optimization skills.
- Proficiency in C++ and Python.
Nice to have
- Knowledge of deep learning models in computer vision, natural language processing, and speech recognition.
- Experience developing high-performance custom kernels for generative AI workloads on GPUs or NPUs.
- Experience with sparsity, reduced precision, layer decomposition, and other model- or layer-level optimization techniques.
- Experience with SSE/AVX, NEON, CUDA, or OpenCL and architecture-specific parallel programming.
- Academic background in computer architecture.
Culture & Benefits
- Work on proprietary NPU technology and next-generation AI compute architectures.
- Collaborate across hardware and software through hardware–software co-design.
- On-site work at the R-TOWER office in Seongnam.
Hiring process
- Application review followed by an online interview.
- On-site interview including an assignment.
- Culture-fit interview, compensation discussion, and final offer.
Будьте осторожны: если работодатель просит войти в их систему, используя iCloud/Google, прислать код/пароль, запустить код/ПО, не делайте этого - это мошенники. Обязательно жмите "Пожаловаться" или пишите в поддержку. Подробнее в гайде →