Project Description:

• Join a challenging project focused on the development and optimization of AI operations on advanced GPU architectures. The role involves high-performance code delivery for open-source software and close collaboration with technical experts and partners to enhance GPU libraries.

Responsibilities:

• Design, implement, and optimize AI operations for GPUs
• Profile and analyze performance to ensure efficient AI workloads
• Deliver robust, high-quality open-source software
• Collaborate with internal specialists and external partners to improve GPU libraries

Mandatory Skills Description:

• Strong expertise in C/C++ programming and parallel programming
• Solid experience in software design, debugging, performance analysis, and test design
• Knowledge of GPU computing technologies such as CUDA, HIP, or OpenCL

Nice-to-Have Skills Description:

• · Experience with low-level optimization techniques (e.g., assembly programming, intrinsic)
• · Understanding of Deep Learning frameworks and AI operations
• · Familiarity with GPU, TPU, DSP, or other hardware accelerators
• · Embedded Performance Optimization

Languages:

• English: advanced