Alibaba Group

Job Description

We are Alibaba DAMO Academy, which is dedicated to exploring the unknown through scientific and technological research and innovation. The driving force behind the Academy is the pursuit of the betterment of humanity.

Our current focus areas include advanced research topics in computing, memory/storage, and interconnect technologies that can revolutionize today's computing systems with holistic innovations ranging from system architectures to VLSI designs, to enable new computing capabilities for improving energy efficiency and performance across multiple application domains, including both high-performance and embedded computing.

We are looking for strong candidates in algorithm/software/hardware co-design for machine learning. In particular, the machine learning frameworks, the compilers for machine learning, the systems for machine learning, and the hardware accelerator modeling for machine learning.

We are looking for candidates experienced in at least two of the three layers of the system design: ML algorithm, ML software implementation, and ML hardware architecture, and co-design those layers.

Responsibilities:

• Research novel ML algorithms.
• Research and develop novel ML operators and integrate into existing ML frameworks.
• Research new ML frameworks and compiler technologies.
• Define and design novel hardware architectures for ML applications.
• Evaluate and optimize performance and quality on the ML algorithms and software on existing and new hardware.
• Develop tools to model new ML accelerators.
• Collaborate with business units who deploy the solutions

Job requirements

• PhD degree in CS/EE/CE or equivalent, minimum 2 years experiences, with strong background on compiler technology
• Experience on Deep-Learning/NN
• Experience on hardware architecture design
• Experience on DL frameworks, such as TensorFlow/Pytorch/TVM, is good plus.
• Strong C/C++ programming skills and unix/linux experience is required, Python programming skills is required.
• Experience on parallel programming such as CUDA/OpenCL/OpenMP is good plus.
• Experience on system performance tuning, performance analysis is a good plus.
• Good written and spoken English.
• Self-motivated, team work, and good communication are a must.