About this role
The team is seeking a GPU Programming Software Engineer on a contract basis to focus on designing, implementing, and optimizing GPU-based software for large language model (LLM) training tasks. This role is ideal for an expert-level professional who is output-driven and can deliver high-quality GPU code that meets specific project requirements.
Key Responsibilities:
- Design GPU-based software solutions and architectures tailored for LLM training workloads.
- Implement and optimize GPU code using technologies such as CUDA, WebGPU, or GLSL as per task specifications.
- Profile and fine-tune GPU kernels and shaders to maximize performance and efficiency.
- Develop host-side logic and integrations in C++ to support GPU execution and tooling.
- Deliver output-based task work that aligns with project specifications and acceptance criteria.
Required Skills & Qualifications:
- Advanced GPU programming experience, particularly targeting NVIDIA chips (CUDA, WebGPU, or GLSL).
- Strong proficiency in C++ for host-side development and integrations.
- Proven experience in profiling and optimizing GPU kernels or shaders for enhanced performance.
- Background in graphics programming, machine learning acceleration, high-performance computing, or related GPU-focused domains.
- Ability to produce task-complete GPU programming work that meets defined specifications.
- Expert-level experience with a quick capacity to engage in productive task work post-onboarding.
Experience:
- Minimum of 5 to 8 years of relevant experience in GPU programming and software engineering.
What we offer:
- A flexible, remote working environment with part-time commitment of 20+ hours per week.
- Opportunity to work on cutting-edge AI development tasks.
- The chance to contribute to impactful projects in the AI training field.
This role is managed by VettedBench on behalf of our client. Your application is reviewed directly by our talent team.