AI Compiler Researcher

We are currently looking for Researchers for our new Computer Architecture Innovation Lab.

What You’ll Do

• NPU-Centric Python Framework Design: Design smart cross-layer optimizations for compilers, runtimes, and frameworks (e.g., PyTorch, vLLM) to boost NPU efficiency. Work on automated model conversion, quantization, and adaptive deployment for NPU workloads.
• Compiler & Toolchain Innovation: Build and improve NPU-specific compiler stacks (LLVM/TVM/GCC) to turn high-level models into super-fast NPU instructions. Optimize scheduling, memory allocation, and parallel execution strategies.
• Hardware-Software Co-Design: Work closely with hardware teams to define NPU ISA extensions, microarchitecture features, and performance counters for maximum efficiency.
• Research & Ecosystem Leadership: Publish your work in top conferences (ISCA, ASPLOS, MLSys) and contribute to open-source projects. Help developers use NPU frameworks through tools, documentation, and partnerships.

Who We’re Looking For

• Master’s degree + 3 years of system software experience or fresh PhD graduates in Computer Science, Computer Engineering, or related fields.
• Strong knowledge of compiler frameworks (LLVM, GCC, TVM, XLA) and NPU/GPU architectures.
• Proficiency in Python and C/C++ for system-level programming.
• Familiarity with AI frameworks (PyTorch, TensorFlow) and model optimization is a plus.
• Experience with xPU/DSP toolchains is a plus.
• Excellent English communication skills.

#J-18808-Ljbffr