The Software Engineer will be part of the software design task force for Artificial Intelligence (AI) related IP. The engineer will be expected to
Enable AI algorithms, models and applications on AI related IP products.
Design solutions for typical AI tech usage scenarios (e.g. AIoT, mobile, AIPC, cloud, and etc), achieving competitive performance and aligning accuracy.
Focus on leading edge algorithm information and implement.
Focus on algorithmic hardware-software co-design for next generation AI related IPs.
Desirable Skills and Experience:
Expert programming skills in C/C++, Python; experience in CUDA/cuTile/Tilelang/Triton is a plus.
Familiar with the underlying implementation of the algorithm, software stack and hardware pipeline.
Familiar with common AI algorithm acceleration methods such as quantization, pruning, sparsification and compression.
Hands-on skill of software integration during all cycles of development.
Experience in implementing algorithms on edge AI devices and know well with typical arch of AI is preferred.
Strong intrinsic interest and self-motivation in the evolution of AI model algorithms and hardware-software co-design, with a certain academic research background; have published papers in top conferences or journals is preferred.
Proficient in training various types of AI models, with preference given to those experienced in multi-node parallel training of large models.
#J-18808-Ljbffr