Skill diminta
CommunicationEnglishPyTorchTroubleshooting
Deskripsi
Deskripsi pekerjaan NVIDIA AI SOFTWARE ENGINEER PT Digital Tech Asia
NVIDIA AI SOFTWARE ENGINEER
- Company: Digital Tech Asia
- Location: Jakarta, Indonesia (Onsite)
- Employment Type: Contract (6–12 Months, Extendable)
- Experience: Minimum 5 Years
- Language: English (Required)
- Availability: Immediate Joiner Preferred
- Job Overview
- Digital Tech Asia is looking for an experienced NVIDIA AI Software Engineer to join an AI Compute / Data Center project.
The ideal candidate will have strong hands-on experience in NVIDIA AI software, GPU computing, and AI inference infrastructure. You will be responsible for building, configuring, optimizing, and troubleshooting AI inference stacks across local GPU environments and cloud-based infrastructure.
Candidates should have practical experience with NVIDIA technologies, including CUDA, NVIDIA NeMo, NVIDIA NIM, and modern AI execution frameworks.
- Key Responsibilities
- Design, build, configure, and optimize AI inference stacks for NVIDIA RTX, DGX, and multi-GPU cloud environments.
- Implement and optimize AI workloads using NVIDIA CUDA, TensorRT, NeMo, and NVIDIA NIM.
- Work with modern AI inference and execution frameworks, including PyTorch, vLLM, Llama.cpp, and TensorRT.
- Optimize GPU utilization, inference performance, latency, and throughput.
- Implement and support topology-aware scheduling tools for GPU workloads.
- Troubleshoot GPU computing, AI software, and inference performance issues.
- Support deployment, integration, and optimization of AI models across local and cloud-based environments.
- Collaborate with infrastructure and AI engineering teams to improve AI Compute platform performance.
- Provide technical expertise and documentation for NVIDIA AI software and tooling.
- Requirements & Qualifications
- Minimum 5 years of experience in AI Software Engineering, GPU Computing, Machine Learning Infrastructure, or a related field.
- Strong hands-on experience building and optimizing AI inference stacks.
- Practical experience with NVIDIA GPU platforms, particularly RTX, DGX, or multi-GPU cloud environments.
- Strong knowledge of NVIDIA CUDA and GPU computing.
- Experience with NVIDIA NeMo frameworks and NVIDIA NIM.
- Hands-on experience with PyTorch, vLLM, Llama.cpp, and TensorRT.
- Understanding of GPU architecture, multi-GPU execution, and inference optimization.
- Experience with topology-aware scheduling or GPU workload scheduling tools.