Senior AI Inference Engineer: Vulkan & Mobile GPUs

Tether Operations Limited; Johannesburg; Remote; ZAR 800 000 - 1 200 000; Full time

Job summary

A leading fintech company is seeking an experienced AI Model Engineer to drive innovations in language model inference and optimization. This role demands expertise in GPU acceleration, particularly with Vulkan, and a solid grasp of quantization methods. The ideal candidate will collaborate globally and push the boundaries of AI technology, contributing to the development of cutting-edge fintech solutions.

Qualifications

• Expertise in kernel development and model optimization.
• Experience with Vulkan for GPU acceleration.
• Strong understanding of quantization techniques.

Responsibilities

• Implement and optimize inference and fine-tuning kernels for language models.
• Design and optimize Vulkan compute shaders.
• Collaborate with teams to integrate optimized models into production.

Skills

Proficiency in C++

GPU kernel programming

GPU acceleration with Vulkan framework

Quantization and mixed-precision optimization

Understanding of LoRA fine-tuning

Debugging GPU performance issues

Mobile GPU acceleration

Large language model architectures

Job description

A leading fintech company is seeking an experienced AI Model Engineer to drive innovations in language model inference and optimization. This role demands expertise in GPU acceleration, particularly with Vulkan, and a solid grasp of quantization methods. The ideal candidate will collaborate globally and push the boundaries of AI technology, contributing to the development of cutting-edge fintech solutions.