below marketmost similar roles pay hereabove market
This role pays more than
56%
of similar roles. Most pay
$144,350–$217,625
— the blue band above.
At the midpoint, this role pays about
$180k
versus about
$181k
for comparable roles.
Based on 240 similar postings.
Employer
About Accenture
Accenture is a leading global professional services company specializing in IT, strategy, consulting, and operations, with a strong focus on digital transformation, cloud computing, and artificial intelligence.
Accenture currently has
174 open roles
on FindRole.
Listed pay typically runs
$94,400–$266,300
across 128 roles with salary data.
The AI & HPC Infrastructure Engineer joins the Global AI Infrastructure team to design, build, and operate advanced infrastructure for AI platforms, GPU-accelerated workloads, and large-scale models. You will deploy and manage XPU-based clusters across bare-metal and containerized environments, providing services like GPUaaS and model serving. Key responsibilities include integrating infrastructure with data pipelines, designing agentic AI systems with secure APIs, and building MCP servers for automated troubleshooting. You will utilize a technical stack including Kubernetes, Slurm, Run:ai, Python, Terraform, and Ansible. Expertise in NVIDIA tools like CUDA, NCCL, and TensorRT-LLM is required, alongside experience with InfiniBand, NVMe, and storage systems like VAST or Weka. This role solves complex performance, scalability, and governance challenges for mission-critical enterprise AI and high-performance computing workloads.
Design and implement AI infrastructure and accelerated computing solutions tailored to specific performance and governance needs.
Deploy and manage XPU-based clusters across bare-metal and containerized environments using Slurm, Run:ai, and Kubernetes.
Integrate AI platforms with existing IT systems, data pipelines, security frameworks, and enterprise governance controls.
Build agentic AI infrastructure by integrating model endpoints, retrieval patterns, and workflow orchestration via secure APIs.
Develop MCP servers and connectors to enable agents to monitor, troubleshoot, and tune infrastructure for high availability.
Architect and deploy NVIDIA platform tools to profile and validate performance across compute and networking layers.
Provide technical guidance and optimization for large-scale training, inference, and multi-node simulation workloads.
Create and maintain technical documentation including architecture diagrams, configuration baselines, and operational runbooks.
What we're looking for
Minimum of 5 years of experience designing, deploying, and managing AI infrastructure and accelerated computing environments across on-premises, cloud, and hybrid environments.
Minimum of 5 years of hands-on experience with accelerated computing platforms including GPUs, DPUs, LPUs, CPUs, high-speed interconnects, and AI storage architectures.
Minimum of 5 years of experience with cluster management, workload scheduling, orchestration, observability, and infrastructure automation using tools like Kubernetes, Slurm, Python, Terraform, and Ansible.
Bachelor's degree or equivalent (minimum 12 years) work experience; if an Associate's Degree is held, a minimum of 6 years of work experience is required.
2+ years of experience implementing MLOps, LLMOps, agentic AI, and DevSecOps frameworks to enable secure, automated, and governed AI workflows (preferred).
2+ years of experience developing APIs, integration services, automation workflows, or platform services using Python and modern API patterns (preferred).
Experience designing agentic AI infrastructure, including LLM inference, RAG, and tool/function calling (preferred).
Experience with NVIDIA platform tools, machine learning frameworks like PyTorch or JAX, and industry certifications in cloud or networking (preferred).