On behalf of Huawei, a world-renowned information and communication technology company, we are seeking passionate and talented individuals to join our team as Architectural Computing Algorithm Engineer

Responsibilities:

Focus on the training and inference optimization of foundation models (LLM/VLM), low-bit quantization techniques, operator fusion and kernel-level optimization, as well as the research and development of the next generation of Spatial Foundation Models.

LLM/VLM Acceleration and Optimization:

• Focus on efficient training and inference acceleration of LLM/VLM models, including but not limited to mixed-precision training, KV cache management, and the deployment of low-bit quantization techniques;
• Continuously track the evolution of basic model operators (such as flash-attention) and emerging algorithmic architectures and computational paradigms.

Spatial Foundation Model Algorithm R&D:

• Participate in the development of spatial foundation models, focusing on 3D scene generation and reconstruction, 3D representation learning, semantic understanding, including but not limited to 3D VAE, 3D Points Cloud Processing, NeRF/3DGS representation, and Scene Graph approaches, as well as generative model techniques such as Diffusion and Autoregressive models;
• Deploy in applications such as autonomous driving and embodied AI. Conduct research on 3D/4D tokenizer for 3D/4D spatial understanding. Design and develop architectures for VLM and VLA foundation models to support spatial understanding, planning, decision-making, and interaction.
• Promote the development and application of next-generation AI systems.

Operator Fusion and Low-Level Kernel Optimization:

• Focus on operator fusion and kernel-level optimization of core components in foundation models;
• Conduct prototyping and optimization of custom kernels targeting diverse hardware platforms such as GPU and NPU, enabling efficient deployment

Qualifications:

• PhD in Computer Science, Mathematics, or related fields; At least 2 years of software development experience with the knowledge of common data structures and techniques, heterogeneous computing and computer architecture.
• Understanding of the basic principles of artificial neural networks, with a background in traditional machine learning. Familiarity with cutting-edge AI algorithms, including but not limited to NLP/LLM/AIGC-related tasks and algorithms.
• Understand and have practical experience with VLM and VLA models or 3D models. Be familiar with mainstream 3D representations and modeling methods such as 3D Gaussian Splatting (3DGS), NeRF, Signed Distance Fields (SDF), and implicit mesh representations. Have knowledge of BEV perception and spatial modeling in autonomous driving, as well as world models and simulation-driven approaches.
• Strong research and analysis skills in algorithms, capable of analyzing and replicating leading conference papers.
• More than 1 year of experience in software development or design, familiar with software processes, engineering methodologies, and tools.
• Proficient in programming languages like Python/C/C++, familiar with various AI frameworks such as PyTorch, TensorFlow, MindSpore, and experienced in using deep learning frameworks for LLM/NLP/AIGC-related content generation and optimization.
• Strong research and analysis skills in algorithms, capable of analyzing and replicating top conference publications, especially in the LLM/NLP/AIGC domain.
• Ability to independently complete the design, development, and self-testing of algorithm modules, meeting company reliability requirements, especially in ensuring the reliability of generated content quality.
• Experience in technical cooperation project management is preferred.