Senior Solutions Architect

E2E Networks Limited; Delhi; On-site; INR 400,000 - 700,000; Full time

Job summary

E2E Networks Limited in Delhi leads complex infrastructure projects combining GPU-accelerated nodes with fast storage and robust networking. This role focuses on end-to-end architecture and advisory for private and public clouds in the Indian market.

You will present designs to C-suite, craft high-level and low-level designs, and guide PoCs from hardware to orchestration with Kubernetes and Docker. Collaboration with sales for TCO and proposals is essential.

Qualifications

• Expertise in Linux systems administration, kernel tuning, and CLI management.
• Strong networking and security knowledge, including VPCs, subnets, firewalls, and high-speed data transfer.
• Experience with virtualization and orchestration (KVM, VMware, Kubernetes).
• Familiarity with GPU architectures and AI frameworks (CUDA, TensorRT, PyTorch, TensorFlow).

Responsibilities

• Lead end-to-end solution architecture for scalable, secure infra with GPU nodes.
• Document High-Level & Low-Level Design for executive presentations.
• Advise clients on data center strategies and cloud architectures for India.
• Drive PoCs for AI model training and inference across stacks.
• Collaborate with sales to build detailed TCO and commercial proposals.

Skills

Linux

Networking

Security

Kubernetes

VMware

Docker

CUDA

PyTorch

TensorFlow

Ceph

Slurm

Tools

KVM

VMware

Docker

Kubernetes

Job description

• End-to-End Solution Architecture: Design scalable, secure, and high-availability

infrastructure. This includes integrating GPU-accelerated nodes with standard

CPU-based workloads, high-speed storage (NVMe/SSD), and complex networking.

• Expert Documentation & Presentation: Lead the creation of professional High-Level Design (HLD) and Low-Level Design (LLD) documents. You must be able to present these to C-suite executives, simplifying complex tech into business value.
• Infrastructure Strategy: Conduct Data Center-level technical assessments. Advise clients on Private vs. Public vs. Sovereign Cloud architectures, focusing on dataresidency and latency for the Indian market.
• The "AI-First" Edge: Lead Proof-of-Concepts (PoCs) for AI model training and inference. You will guide clients on optimizing their stack—from the hardware layer up to the orchestration layer (Kubernetes/Docker).
• TCO & Proposal Engineering: Collaborate with Sales to build detailed commercial proposals. You must be able to justify the Total Cost of Ownership (TCO) of E2E’s specialized infra vs. generic hyperscaler offerings.

Technical Qualifications

• Expertise in Linux Systems: Deep knowledge of Ubuntu/CentOS/Debian environments, kernel tuning, and CLI-based management.
• Networking & Security: Strong understanding of VPCs, Subnetting, Firewalls, Load Balancers, and RDMA/InfiniBand for high-speed data transfer.
• Storage Tiers: Proficiency in architecting Block, Object, and File storage solutions for different performance tiers.
• Virtualization & Orchestration: Hands-on experience with KVM, VMware, and heavy expertise in Kubernetes for containerized workloads.
• Accelerated Computing: Understanding of NVIDIA’s GPU architecture (Hopper/Ampere/Ada Lovelace) and how to match specific SKUs (e.g., A100 vs. L4s) to customer workloads.
• AI Stack Knowledge: Familiarity with the NVIDIA AI Enterprise (NVAIE) suite and common frameworks (PyTorch, TensorFlow).
• GPU Cloud Architecture & Provisioning: Deep understanding of provisioning GPU instances across bare-metal and virtualized environments.
• Multi-GPU Orchestration: Experience with distributed training and inference.
• GPU Resource Scheduling: Familiarity with Slurm, KubeFlow, or Ray Cluster for task scheduling, job management, and workload optimization.
• Inference Optimization: Hands-on understanding of TensorRT, ONNX Runtime, and CUDA/cuDNN tuning for low-latency inference pipelines.
• torage & Data Pipeline Integration: Ability to design AI data pipelines that feed high-throughput training workflows—experience with CVAT/DALI, Ceph, or NFS-over-RDMA.
• Ecosystem Familiarity: Exposure to NVIDIA DGX, HGX, or cloud-native GPU platforms. Soft Skills & "The Face of E2E"