Senior AMD GPU Performance Engineer for vLLM (Singapore)

Inferact; Singapore; On-site; SGD 200,000 - 400,000; Full time

Benefits offered by this job

Medical coverage

Dental coverage

Vision coverage

Equity options

Job summary

Inferact is hiring an AMD GPU performance engineer in Singapore to enhance vLLM for AMD accelerators. You'll develop and optimize AMD GPU backends using tools like ROCm and Triton to ensure top-tier performance.

Ideal candidates have a Bachelor's degree in a related field and hands-on experience with AMD GPU workloads. The role offers a competitive salary range of SGD 200,000 to 400,000 annually, along with comprehensive benefits including medical, dental, and vision coverage.

Qualifications

• Bachelor's degree or equivalent experience in computer science or engineering.
• Hands-on experience optimizing AMD GPU workloads using ROCm or similar tools.
• Deep understanding of AMD GPU execution, memory behavior, and kernel performance.
• Experience optimizing ML kernels such as attention or GEMM.
• Strong benchmarking skills with hardware counters and reproducible benchmarks.

Responsibilities

• Build and optimize AMD GPU backends and kernels for vLLM.
• Work on performance-critical paths such as attention and GEMM.
• Deliver fast, benchmarked, and maintainable AMD GPU support in vLLM.

Skills

AMD GPU optimization

Performance profiling

Benchmarking

Machine learning

Education

Bachelor's degree in computer science or engineering

Tools

ROCm

HIP

Triton

CK

AITER

Job description

Inferact is hiring an AMD GPU performance engineer in Singapore to enhance vLLM for AMD accelerators. You'll develop and optimize AMD GPU backends using tools like ROCm and Triton to ensure top-tier performance.

Ideal candidates have a Bachelor's degree in a related field and hands-on experience with AMD GPU workloads. The role offers a competitive salary range of SGD 200,000 to 400,000 annually, along with comprehensive benefits including medical, dental, and vision coverage.