Hardware Engineer (Remote | $80 –$100/hr)

Synthires; India; On-site; INR 10,526,000 - 13,158,000; Part time

Job summary

Synthires in India seeks experienced GPU Performance Engineers, CUDA Developers, and GPU Kernel Optimization Specialists to advance AI research and evaluation projects. You will analyze, optimize, and evaluate GPU kernels across modern hardware architectures using CUDA, C++17, profiling, and performance tuning.

You will collaborate with teams to improve AI-related GPU workloads, document methodologies and profiling results, and ensure efficient hardware utilization.

Qualifications

• Availability to work at least 20 hours per week.
• Strong proficiency in C++ through C++17.
• Working knowledge of Python and Git.
• Experience with at least one GPU programming framework (CUDA, HIP, Slang, HLSL, GLSL).
• 1+ year of GPU programming or optimization research.
• Strong understanding of GPU architecture and profiling techniques.
• Experience using profiler metrics to optimize GPU kernels.

Responsibilities

• Analyze and optimize GPU kernels for performance, efficiency, scalability, and hardware utilization.
• Use profiler metrics (L2 cache, occupancy, memory throughput, warp efficiency) to guide optimization.
• Identify bottlenecks and recommend improvements.
• Develop, review, and optimize C++17, Python, and GPU code.
• Apply CUDA, HIP, shader programming to improve kernel performance.
• Document optimization methodologies, profiling results, and decisions with clear reasoning.
• Collaborate with engineering teams to evaluate and improve AI-related GPU workloads.

Skills

CUDA

C++17

Python

Git

GPU profiling

Performance optimization

Tools

NVIDIA Nsight Compute

HIP

Slang

HLSL

Job description

This opportunity is for experienced GPU Performance Engineers, CUDA Developers, and GPU Kernel Optimization Specialists interested in contributing to advanced AI research and evaluation projects.

The role focuses on analyzing, optimizing, and evaluating GPU kernels across modern hardware architectures. You'll leverage your expertise in CUDA, C++17, GPU profiling, and performance optimization to improve computational efficiency and help advance next-generation AI systems.

This is a contract-based opportunity for professionals passionate about maximizing GPU performance and hardware utilization.

Responsibilities

• Analyze and optimize GPU kernels for performance, efficiency, scalability, and hardware utilization.
• Use profiler metrics such as L2 cache hit rate, occupancy, memory throughput, warp efficiency, and related performance indicators to guide optimization decisions.
• Identify bottlenecks in GPU kernel implementations and recommend performance improvements.
• Develop, review, and optimize C++17, Python, and GPU programming code.
• Apply expertise in CUDA, HIP, shader programming, or related GPU programming frameworks to improve kernel performance.
• Document optimization methodologies, profiling results, and engineering decisions with clear technical reasoning.
• Collaborate with engineering teams to evaluate and improve AI-related GPU workloads.

Required Qualifications

• Availability to work at least 20 hours per week.
• Strong proficiency in C++ (through C++17).
• Working knowledge of Python and Git.
• Professional experience with at least one GPU programming framework, including CUDA, HIP, Slang, HLSL, GLSL, or similar technologies.
• 1+ year of professional or graduate-level research experience working with GPU programming or optimization.
• Strong understanding of GPU architecture and performance profiling techniques.
• Experience using profiler metrics to optimize GPU kernels efficiently.
• Strong analytical and problem-solving skills with excellent attention to performance optimization.

Preferred Qualifications

• Experience with CUDA C++ Core Libraries, inline PTX assembly, or Tensor Core optimization.
• Experience optimizing kernels for NVIDIA Blackwell or other modern GPU architectures.
• Familiarity with NVIDIA Nsight Compute or similar GPU profiling tools.
• Experience working with GPU platforms from NVIDIA, AMD, Qualcomm, or related hardware vendors.
• Contributions to open-source GPU optimization or high-performance computing projects.
• Experience supporting AI, machine learning, or high-performance computing workloads.

Compensation

• Competitive compensation of $80–$100/hour.
• Weekly payments.