CUDA Engineer

This is a fully remote position, open to applicants in Romania.

📋 Description

• Analyze and enhance the performance of LLM inference on contemporary GPUs.

• Collaborate with frameworks for inference and serving, including vLLM, SGLang, TensorRT-LLM, or other comparable systems.

• Examine GPU execution, memory access patterns, kernel efficiency, and identify inference bottlenecks.

• Create, refine, debug, or reverse-engineer CUDA kernels and low-level GPU tasks.

• Investigate optimization strategies for matrix multiplication, attention mechanisms, and neural network layers.

• Conduct performance benchmarks and discover areas for improving latency, throughput, and GPU utilization.

• Engage in technically demanding AI systems and high-performance computing challenges.

• Carry out practical experiments and convey technical insights effectively.

⛳️ Requirements

• Extensive hands-on experience with CUDA and GPU programming.

• Profound knowledge of GPU architecture, memory hierarchy, parallel execution, and performance enhancement techniques.

• Experience in LLM inference, AI systems, GPU kernels, or high-performance computing.

• Familiarity with vLLM, SGLang, TensorRT-LLM, Triton, or FlashAttention.

• Proven ability to debug intricate performance challenges and investigate low-level system behavior.

• Strong analytical and problem-solving abilities.

• Capacity to articulate complex technical ideas in English.

• A PhD or research experience in Computer Science, Machine Learning, Computer Architecture, or a related discipline (preferred).

• Contributions to publications, research projects, or significant engagement with open-source GPU/LLM infrastructure (preferred).

• Experience with PyTorch internals, distributed inference, quantization, KV cache optimization, or custom GPU kernels (preferred).

• Contributions to vLLM, SGLang, Triton, FlashAttention, or similar repositories (preferred).

🏝️ Benefits

• Security through thorough client vetting and prompt, reliable payments.

• Career support and assistance in discovering new project opportunities.

• Legal guidance for independent contractor or sole proprietorship status, tax matters, and related processes.

• English language courses.

• Opportunities for professional development.

• Team-building activities.

• People-first management style with minimal bureaucracy.

• A welcoming company culture.

• Flexible working hours.

• 29 days of paid time off (18 working days per year plus all national holidays).

• 10 paid recovery days.

• Comprehensive financial and legal support for independent contractors.

• Free English classes with native speakers or Ukrainian instructors.

• Dedicated HR support.