NVIDIA AI Software Engineer
Description
Company: FPT Software
Location: Jakarta, Indonesia (hybrid; on-site client collaboration as needed)
Employment type: Full-time
About FPT Software
FPT Software is a global technology and IT services company headquartered in Vietnam, delivering digital transformation, AI, and engineering solutions to clients across Asia-Pacific, Europe, Japan, and North America. In Jakarta, our team partners with enterprises on building and deploying production-grade AI systems powered by the NVIDIA ecosystem.
Role Overview
We are looking for an AI Software Engineer with strong NVIDIA stack experience to design, optimize, and deploy AI solutions on GPU infrastructure. You will work with cross-functional teams and clients to turn models into scalable, low-latency, cost-efficient products, from generative AI and LLM applications to computer vision and edge deployments.
Key Responsibilities
- Design and develop AI applications and services using NVIDIA technologies such as CUDA, cuDNN, TensorRT, TensorRT-LLM, Triton Inference Server, NeMo, and NIM microservices.
- Optimize model training and inference performance through quantization, batching, kernel-level tuning, and GPU memory management.
- Build and deploy LLM / RAG / agentic solutions, including fine-tuning, evaluation, guardrails (e.g., NeMo Guardrails), and serving at scale.
- Develop computer vision and multimodal pipelines using DeepStream, Metropolis, or similar frameworks.
- Containerize and deploy AI workloads on Kubernetes and cloud or on-prem GPU clusters (NGC containers, Docker, Helm, CI/CD).
- Profile and debug GPU workloads using tools such as Nsight Systems, Nsight Compute, and DCGM.
- Collaborate with solution architects, data scientists, and client stakeholders to translate business requirements into technical designs.
- Prepare proofs of concept, benchmarks, technical documentation, and client demos.
- Stay current with NVIDIA's software ecosystem and share knowledge through internal sessions and mentoring.
Requirements
Must have
- Bachelor's or Master's degree in Computer Science, AI, Engineering, or a related field.
- 3+ years of software engineering experience, with at least 2 years in AI/ML engineering.
- Strong proficiency in Python and good working knowledge of C++.
- Hands-on experience with deep learning frameworks (PyTorch and/or TensorFlow).
- Practical experience with GPU-accelerated AI development and at least one of: TensorRT, Triton Inference Server, CUDA, NeMo.
- Experience deploying ML models to production using Docker and Kubernetes.
- Solid understanding of Linux environments, Git, and CI/CD practices.
- Good communication skills in English; Bahasa Indonesia is a strong advantage.
Nice to have
- Experience with LLM inference optimization (vLLM, TensorRT-LLM, KV-cache, speculative decoding).
- Knowledge of distributed training (multi-GPU / multi-node, NCCL, DeepSpeed, Megatron).
- Experience with RAPIDS, DeepStream, Jetson / edge AI, or Omniverse.
- Familiarity with cloud platforms (AWS, Azure, GCP) and MLOps tools (MLflow, Kubeflow).
- NVIDIA Deep Learning Institute (DLI) or NVIDIA-Certified Associate/Professional credentials.
- Prior experience in client-facing or IT outsourcing/consulting environments.
What We Offer
- Competitive salary and performance-based bonus.
- Opportunities to work on cutting-edge AI projects with global clients.
- Training and certification support, including NVIDIA-related learning paths.
- Career growth within a large international technology organization.
- Health insurance and other benefits in line with Indonesian regulations and company policy.
- Collaborative, multicultural team environment.
Skills
Sign in to read and join the discussion.