Machine Learning Engineer - AI Compiler Optimization

What you'll do

• Implement graph-, operator-, and memory-level compiler optimizations, including graph-operator fusion and automatic operator generation.
• Collaborate with hardware and algorithm teams to optimize compilation strategies for hardware characteristics.
• Adapt PyTorch recommendation models to the engine by optimizing model import, conversion, and code generation.

What they ask for

• Proficiency in a mainstream AI compiler framework such as Triton, MLIR, or TVM, with hands-on customized compilation optimization and Pass development experience.
• GPU/NPU compilation optimization experience, including loop, memory, and operator optimization, and independent performance bottleneck analysis.
• Familiarity with common deep learning model structures and compilation adaptation in PyTorch and TensorFlow.

Nice to have

• Recommendation-engine architecture and implementation experience in compilation and low-latency inference optimization for large-scale, high-concurrency recommendation systems.
• Contributions to open-source AI compiler projects such as TVM or MLIR, or expertise in large-model compilation adaptation for recommendation and sparse operator generation.

Tools

Triton, MLIR, TVM, GPU, NPU, PyTorch, TensorFlow

AI focus

Optimizes AI compilation and inference for recommendation models, including compiler passes, hardware-aware execution, model adaptation, and operator generation.