We are looking for the best computing scientists, architects, and senior engineers. We are building industry-leading AI computing accelerating engine and platform, building the most cutting–edge artificial intelligence and applied machine learning algorithms and computing-systems which combined inference and training software and hardware to apply in the largest AI datacenter in the world. We are managing and optimizing artificial intelligence and machine learning algorithms (for Cloud Computing, Search, Ads, News Feed, Autonomous Driving, Photos, Computer Vision, Speech, NLP, etc), improving computing efficiency, defining the computing-systems and platform design. Responsibilities: 1. Participate and lead Inference Engine framwork development and operator level optimization. 2. Develop and optimize AI/ML computing commnunication libraries and algorithms. 3. Develop and optimize compiler system, libraries, and algorithms for CPU/GPU/FPGA/ASIC etc. 4. Participate and lead high performance computing platform design and development.
1. MS or PhD in Computer Science related Machine Learning field or equivalent practical experience. 2. Excellent programming experience in one or more of the following: C, C++, Python. 3. Applied or develop experience on AI/ML algorithm and major framewoks(Tensorflow, Caffe, Mxnet, PaddlePaddle, etc.) 4. Professional experience with comiler development on GPU/ARM/MIPS/DSP etc. 5. Professional experience with communication libraries for large scale distributed systems or parallel systems (MPI, NCCL, RCCL, RDMA, etc.) Preferred qualifications: - Well understand major framework Caffe/Caffe2/Mxnet/TensorFlow/PaddlePaddle, with good experience on source code optimization and compiling. - Excellent programming experience with cublas、cudnn、MIopen、OpenBlas、MKL、Eigen etc. - Experience with CUDA/OpenCL. - Experience with Neon, ARM-GPU development and optimization . - Experience with performance analyze tools, CodeXLNVVPGPA etc. - Experience with LLVM. - Experience with Linux Kernel develoment and optimization.