NVIDIA AI Platform Engineer (CUDA/PyCUDA)

Job Summary

We are seeking a Senior NVIDIA AI Platform Engineer to design, develop, and optimize enterprise AI applications using NVIDIA AI technologies. This role focuses on building GPU-accelerated AI solutions, integrating NVIDIA AI Enterprise components, deploying and optimizing NVIDIA open models, and packaging enterprise AI applications for deployment. You will work on CUDA/PyCUDA development, TensorRT optimization, NVIDIA NIM, Triton Inference Server, and AI platform engineering to deliver high-performance AI solutions for enterprise customers.

Key Responsibilities

• Design and develop GPU-accelerated AI applications using CUDA and PyCUDA.
• Deploy, optimize, and integrate NVIDIA open models and NVIDIA AI Enterprise technologies (NIM, NeMo, TensorRT, Triton Inference Server).
• Build scalable AI APIs and inference services using Python and modern backend frameworks.
• Optimize AI model inference, GPU utilization, and multi-GPU performance for enterprise workloads.
• Develop reusable AI components and package enterprise AI solutions for deployment through the NVIDIA AI ecosystem and enterprise marketplaces where applicable.
• Collaborate with AI, product, and engineering teams to build enterprise-grade AI platforms and intelligent agents.
• Ensure high availability, performance, security, and scalability of AI infrastructure and GPU-based applications.

Required Skills

• CUDA, PyCUDA, Python
• NVIDIA NIM, TensorRT, Triton Inference Server, NeMo
• Experience deploying and optimizing LLMs and NVIDIA/open-source foundation models
• Docker, Kubernetes, Linux, GPU infrastructure
• REST APIs, FastAPI, AI platform development
• Experience with AI frameworks such as LangChain, LangGraph, or similar is a plus
• Strong software engineering, performance optimization, and distributed AI systems knowledge