Remote | CUDA Engineering Expert — $60–$100/hour
About The Position
We are sharing a specialised consulting opportunity for experienced CUDA Engineering Experts with strong expertise in CUDA, C++, GPU kernel optimisation, GLSL, WebGPU, profiling, and high-performance computing to contribute to an advanced GPU-engineering and AI training project. Selected professionals will analyse and optimise GPU kernels, improve CUDA and C++ codebases, work with shader and WebGPU workflows, and document performance improvements across modern GPU architectures. The work requires strong low-level performance engineering, rigorous profiling, and the ability to translate technical findings into clear, actionable recommendations. No prior experience in AI is required.
Requirements
- Demonstrated expertise in CUDA programming
- Strong track record of GPU kernel performance optimisation
- Advanced C++ development experience
- Experience working in high-performance computing environments
- Hands-on experience with GLSL and WebGPU
- Strong understanding of graphics or compute shader development
- Proficiency with GPU profiling tools such as Nsight, Visual Profiler, or equivalent
- Ability to reason about memory behaviour, kernel execution, and hardware utilisation
- Strong analytical skills for evaluating performance across GPU architectures
- Experience refactoring performance-critical codebases
- Excellent written and verbal technical communication skills
- Comfortable collaborating in remote, cross-disciplinary environments
Nice To Haves
- No prior AI-training experience is required
Responsibilities
- Analyse and optimise GPU kernels using CUDA
- Identify performance bottlenecks affecting computational throughput
- Improve kernel efficiency across modern GPU hardware
- Apply targeted optimisation techniques based on profiling results
- Validate performance gains using quantitative benchmarks
- Profile GPU workloads using tools such as Nsight, Visual Profiler, or comparable platforms
- Diagnose memory, compute, occupancy, and execution bottlenecks
- Evaluate kernel performance across different hardware generations
- Develop data-driven optimisation strategies
- Document measurable changes in latency, throughput, and resource utilisation
- Refactor CUDA and C++ code for improved maintainability and efficiency
- Improve architecture and organisation within performance-critical codebases
- Reduce unnecessary complexity while preserving functionality
- Adapt implementations for portability across GPU architectures
- Apply high-performance computing best practices throughout development
- Implement shader logic using GLSL
- Develop graphics and compute workflows using WebGPU
- Integrate shader-based processing into existing systems
- Evaluate performance trade-offs across GPU execution environments
- Maintain compatibility and consistency across pipeline components
- Contribute expertise to GPU architecture and performance-design discussions
- Evaluate new GPU-based implementation approaches
- Assess technical trade-offs across performance, maintainability, and scalability
- Support definition of meaningful performance metrics
- Recommend practical approaches based on profiling and benchmarking evidence
- Document optimisation strategies, benchmark results, and technical findings
- Produce clear reports describing performance improvements
- Communicate complex GPU behaviour to technical stakeholders
- Collaborate with remote and cross-disciplinary project teams
- Share relevant developments in GPU programming and performance engineering
Benefits
- Independent contractor engagement
- Fully remote
- Compensation: $60–$100/hour
Stand Out From the Crowd
Upload your resume and get instant feedback on how well it matches this job.
What This Job Offers
Career Level
Senior
Education Level
No Education Listed
