AI Researcher

• Conduct original research in distributed and parallel computing for large-scale AI model training and inference.
• Develop novel methods for automatic parallelization, computation graph optimization, and numerical efficiency.
• Collaborate with Huawei's global research and engineering teams to transfer research outcomes into real-world AI systems.
• Publish results in top-tier conferences and journals (e.g., NeurIPS, ICML, ICLR, MLSys, OSDI, SC).
• Contribute to open-source initiatives or community research collaborations in AI infrastructure.

Focus Areas

• Distributed Parallelism: Automatic, pipeline, and tensor parallelism; communication optimization.
• Graph Optimization: Dynamic/static graph scheduling, compiler-based optimization.
• Numerical Computation: Sparse/mixed-precision training, low-rank decomposition, and dimensionality reduction.
• Heterogeneous Computing: Integration across CPUs, GPUs, NPUs, and custom accelerators.

• Ph.D. or equivalent experience in Computer Science, Electrical Engineering, or a related field.
• Strong background in distributed systems, AI infrastructure, compiler optimization, or numerical computing.
• Proven ability to publish in top-tier venues or deliver impactful research results.
• Experience with large-scale AI or HPC platforms preferred.
• Fluent in English; French proficiency is a plus.