Project Role : Technology Platform Engineer

Project Role Description : Creates production and non-production cloud environments using the proper software tools such as a platform for a project or product. Deploys the automation pipeline and automates environment creation and configuration.

Must have skills : Linux Architecture

Valuable to have skills : Machine Learning (ML), Cloud Technology Architecture, Docker Kubernetes Administration, Edge Computing, Microsoft Agentic AI Architect

Minimum 5 Year(s) Of Experience Is Required

Educational Qualification : 15 years full time education

Key Responsibilities

- Design and implement HPC and AI infrastructure solutions, aligning system architecture and deployment roadmaps to industry-specific performance and scalability needs
- Deploy, configure, and manage XPU-based clusters (CPU/GPU/accelerators) using schedulers, VM/K8s orchestration platforms, Slurm, and containerized platforms in scalable designs to provide Metal as a Service (MaaS), GPUaaS, AIaaS, and other offerings
- Optimize cluster performance, scalability, energy, and cost efficiency across on-premises, cloud, and hybrid environments
- Integrate AI and HPC platforms with existing IT systems, data pipelines, and security frameworks
- Monitor, troubleshoot, and tune infrastructure to ensure high availability, low-latency networking, and workload resiliency
- Develop and maintain documentation including architecture diagrams, configuration baselines, and operational runbooks
- Provide Provide technical guidance and support to users, enabling efficient execution of HPC/AI workloads, large-scale models, and simulations





Required Skills and Qualifications
- Experience in enterprise-wide HPC strategy and architecting next-gen supercomputing environments across the full stack.
- Primary Skills: Linux Administrator/Architect, Advanced CUDA/GPU on H100/A100, HPC cluster design (SLURM/PBS Pro)
- Secondary Skills: Parallel programming: MPI, OpenMP, CUDA, SYCL, NVIDIA DGX SuperPOD & BCM, InfiniBand NDR 400Gb/s & RoCEv2 design, MLOps & HPC integration (Kubeflow), Containerization: Singularity, Kubernetes, DevOps: Ansible, Terraform for HPC, UFM management, RunAI, Azure ML integration: distributed training, MLflow, Terraform / Bicep IaC for Azure HPC, FinOps: Reserved Instances, Spot VM strategies, Hybrid cloud HPC: on-prem to Azure/AWS burst
- Proven hands-on experience designing, deploying, and managing HPC and AI infrastructure across on-premises, cloud, and hybrid environments in 2 or more segments: hyperscaler, neocloud, large Enterprise, Telco/Mobile, supporting key industries such as Financial Services, Life Sciences, Manufacturing, and Retail
- Deep knowledge of accelerated computing architectures (GPUs, XPUs, DPUs), high-performance fabrics (InfiniBand, Ethernet), SONiC, networking, and modern storage/data platforms (e.g. NVMe-oF, Lustre, GPFS, BeeGFS, VAST, DDN,



Weka) to build robust solutions
- Proficiency with cluster management and orchestration (e.g. Slurm, Run:ai, Kubernetes, Docker), real-time performance monitoring, and observability frameworks
- Hands-on experience with cloud and virtualization platforms (e.g. AWS, Azure, GCP, VMware, Nutanix) and expertise in automation and optimization using scripting (Python, AI tools) with foundational Infrastructure-as-Code tools such as Terraform and Ansible.

Preferred Skills and Qualifications
- Experience managing the deployment of 1,000+ GPU clusters for HPC and AI workloads with various infrastructure services enabled
- Experience with GPU computing libraries and accelerators (e.g., NVIDIA CUDA, Dynamo, AMD ROCm).
- Experience with AI and HPC Networking (e.g., RoCE, InfiniBand, muti-planar/multi-rail designs, platform buffer architectures)
- Knowledge of Machine Learning and AI frameworks (e.g., TensorFlow, PyTorch, JAX), Jupyter notebooks / Google Colab environments
- Familiarity with DevOps practices and tools (e.g., Ansible, Terraform) for infrastructure automation
- Experience with AgenticAI and associated technologies to leverage and build agents for workflow automation and observability
- Industry certifications in NVIDIA infrastructure, public cloud providers, Data Science, etc. are a plus
- Strong problem-solving, troubleshooting, communication, and collaboration skills to deliver reliable, scalable, and high-performance infrastructure solutions in fast-paced, dynamic environments that reward technical talent

📌 Technology Platform Engineer (Gurugram)
🏢 Accenture in India
📍 Gurugram