GPU Kernels AI Task Auditor - Freelance AI Trainer Project
About this role
Meridial needs technical specialists to resolve the quality and accuracy gaps in AI training data specifically for GPU kernel development. Specialists audit pre-existing evaluation tasks to ensure workflows remain technically rigorous, practical, and fully reproducible for advanced machine learning systems.
Day-to-day work involves reviewing complex scenarios to pinpoint hardware-level performance optimization flaws, execution bottlenecks, and memory allocation bugs. Candidates who succeed here possess deep expertise in parallel computing and CUDA or Triton programming, alongside the analytical rigor required to troubleshoot intricate technical logic.
What you will do
- Evaluate whether GPU kernel tasks are technically accurate, realistic, solvable, and supported by reliable tests.
- Assess existing evaluation criteria to guarantee tasks meet high standards of precision.
- Perform technical audits to identify memory allocation bugs and execution bottlenecks.
- Detect parallel computing logic errors within assigned tasks.
- Provide clear and actionable feedback on all identified technical issues.
What you need
- Demonstrable professional experience and deep technical knowledge in GPU Kernel development.
- Hands-on expertise in CUDA or Triton programming.
- Familiarity with parallel computing and hardware-level performance optimization.
- Strong analytical and problem-solving skills for testing and troubleshooting complex technical scenarios.
Skills and tech stack
Questions candidates ask
What is the remote work policy for this project?
The project operates on a remote, worldwide basis. As a contractor, you will supply your own secure computer and high-speed internet.
How much does the role pay?
The pay rate is $60 per hour. The exact rate is determined after evaluating your experience, expertise, and geographic location.
What benefits are included with the contract?
Company-sponsored benefits such as health insurance and PTO do not apply to this freelance contract. You operate strictly as an independent contractor.
What kind of technical expertise is required?
Candidates need professional experience and deep knowledge in GPU Kernel development, including CUDA or Triton programming. You will also focus on parallel computing and hardware-level performance optimization.