AI Inference Engineer

Must have skills: vLLM, model inference/serving, transformer architecture, GPU workloads, Python. AI/model-serving SME.

Job Nice to Haves: NVIDIA Triton Inference Server, NVIDIA NIMs, GPU-based inference, Model deployment/hosting, Kubernetes, Python/software development, GPU/inference monitoring

About the Position / Current Initiatives: This is the person who deeply understands how AI models are actually hosted and served at scale. They'll work directly with the inference engines and help manage/model the actual serving environment not simply consume OpenAI or other APIs.

Position Summary: This role is crucial for understanding how AI models are hosted and served at scale. The candidate will work directly with inference engines, managing and modeling the serving environment beyond simply consuming APIs from providers like OpenAI. The position requires deep technical expertise, particularly in vLLM, AI/model inference, and transformer-based architecture. While enterprise experience is not mandatory, a strong technical background is essential.

Extensive knowledge of vLLM and AI/model inference

Experience with transformer-based architectures

Familiarity with GPU-based inference and related technologies

Directly manage inference engines and the model serving environment

Monitor and troubleshoot GPU utilization, traffic, and memory

Ensure model performance and address any misbehavior or tier 3 debugging issues

Collaborate on deploying models at scale using technologies like NVIDIA Triton Inference Server and Kubernetes