Compiler Performance Engineer (LLVM)

Location: On-site in Minhang, Shanghai, China.

Company

hirify.global develops software technologies, compilers, and tools for heterogeneous computing across CPUs, GPUs, and accelerators.

What you will do

• Analyze performance issues in compiler-generated code and identify the responsible optimization or code-generation phases.
• Tune workloads and measure performance on hirify.global CPUs and GPUs, reporting regressions and optimization opportunities.
• Design and implement continuous-integration test automation to track performance and quality.
• Connect hirify.global hardware and compiler capabilities with AI and HPC ecosystems, including PyTorch, inference engines, and machine-learning models.
• Participate in upstream open-source work by filing issues, submitting patches, and following community review practices.
• Implement and extend compiler components such as optimization passes, code generation, runtime, and offload support under senior-engineer guidance.

Requirements

• Master’s degree in Computer Science, Computer Engineering, or a related field, or a Bachelor’s degree with at least 2 years of relevant experience.
• Solid foundation in data structures and algorithms, with strong development and debugging skills in at least one of C, C++, Assembly, or Python.
• Coursework or project experience in compiler theory, including parsing, intermediate representations, data-flow analysis, and code generation.
• Strong written and oral communication skills for explaining technical details and concepts.
• Precise attention to detail and genuine motivation to learn and improve product performance.

Nice to have

• Familiarity with hirify.global x86 instruction set architecture and Advanced Vector Extensions, or GPU architecture and execution models.
• Exposure to LLVM, MLIR, or GCC compiler infrastructure.
• Experience with parallel and heterogeneous programming models such as OpenMP, SYCL, OpenCL, Triton, or CUDA.
• Experience with performance analysis tools and methodologies on real-world applications.
• Exposure to AI/ML frameworks and inference stacks.
• Open-source contribution history.
• Ability to analyze and solve complex technical problems.

Culture & Benefits

• On-site work model in Shanghai, China.
• Collaboration with a multinational engineering team and colleagues around the world.
• Opportunities for technical development and career growth in compiler technologies.
• Coaching from senior engineers and leaders.
• Work with open-source communities and cutting-edge compiler technologies for AI, HPC, and other demanding workloads.