Technical SME L3

Vyntrapro Innovations; Indonesia; On-site; IDR 390,600,000 - 669,600,000; Full time

Job summary

Vyntrapro Innovations seeks an AI Software & Inference Engineer to design, optimize, and manage AI model deployment environments across on-premises and cloud GPU infrastructures.

You will drive performance, scalability, and cost efficiency while collaborating with Data Scientists, MLOps, and Infrastructure teams on complex GPU workloads.

Qualifications

• 5+ years of AI/ML engineering or GPU infrastructure experience.
• Deep expertise in NVIDIA GPU software stack and acceleration technologies.
• Hands-on experience with AI inference optimization and model serving.
• Strong Python programming and Linux administration skills.

Responsibilities

• Provide deep technical expertise across NVIDIA AI software platforms (CUDA, NVIDIA NeMo, NIM, TensorRT) and GPU acceleration technologies.
• Design, build, and optimize AI inference platforms for multi-GPU and cloud environments.
• Deploy and fine-tune LLMs and Generative AI workloads for production use cases.
• Develop and optimize inference pipelines using PyTorch, vLLM, Llama.cpp, TensorRT-LLM, and Hugging Face.
• Implement topology-aware scheduling and workload placement to maximize GPU utilization.
• Optimize model serving latency, throughput, and memory usage across clusters.
• Collaborate with Data Scientists, MLOps, and Infrastructure teams to deploy scalable AI solutions.
• Troubleshoot GPU, CUDA, driver, networking, and distributed inference issues.
• Evaluate and adopt NVIDIA ecosystem updates and emerging AI software technologies.
• Establish benchmarking, monitoring, and observability for AI workloads.
• Support multi-node, multi-GPU distributed inference and model serving architectures.
• Create technical documentation and operational best practices.

Skills

Python programming

Linux administration

AI/ML engineering

Tools

CUDA

NVIDIA NeMo

NVIDIA NIM

TensorRT

PyTorch

vLLM

Llama.cpp

TensorRT-LLM

Hugging Face

Job description

Job Summary

We are seeking a highly skilled AI Software & Inference Engineer with strong expertise in NVIDIA AI software ecosystems, GPU acceleration technologies, and large-scale AI inference platforms. The ideal candidate will be responsible for designing, optimizing, and managing AI model deployment environments across on-premises and cloud-based GPU infrastructures, ensuring maximum performance, scalability, and cost efficiency.

Key Responsibilities

• Provide deep technical expertise across NVIDIA AI software platforms, including CUDA, NVIDIA NeMo, NVIDIA NIM, TensorRT, and GPU acceleration technologies.


• Design, build, and optimize AI inference platforms for RTX, DGX, and multi-GPU cloud environments.


• Deploy and fine-tune Large Language Models (LLMs) and Generative AI workloads for production use cases.


• Develop and optimize inference pipelines using modern AI frameworks such as PyTorch, vLLM, Llama.cpp, TensorRT-LLM, and Hugging Face.


• Implement topology-aware scheduling and workload placement strategies to maximize GPU utilization and performance.


• Optimize model serving latency, throughput, memory utilization, and GPU resource allocation.


• Collaborate with Data Scientists, MLOps Engineers, and Infrastructure teams to deploy scalable AI solutions.


• Troubleshoot complex GPU, CUDA, driver, networking, and distributed inference issues.


• Evaluate and implement emerging AI software technologies and NVIDIA ecosystem updates.


• Establish performance benchmarking, monitoring, and observability frameworks for AI workloads.


• Support multi-node, multi-GPU distributed inference and model-serving architectures.


• Create technical documentation, architecture standards, and operational best practices.

Required Qualifications

• 5+ years of experience in AI/ML Engineering, GPU Infrastructure, AI Platform Engineering, or related roles.


• Strong expertise in NVIDIA GPU software stack and acceleration technologies.


• Hands-on experience with AI inference optimization and model serving.


• Strong Python programming and Linux administration skills.


• Experience with containerized environments using Docker and Kubernetes.


• Knowledge of distributed computing, GPU networking, and AI infrastructure architecture.