AI Infrastructure Solutions Architect

Location & work modality: India ( Remote)

About Submer

Submer enables organizations scaling AI to overcome the limits of traditional datacenters across power, compute density and efficiency.

We design and deliver scalable, high-density AI datacenter infrastructure built around industry-leading liquid cooling, supporting everything from early AI deployments to full-scale production environments.

What Impact you will have

We are looking for AI Infrastructure Solutions Architect

Department: Solutions Engineering / Presales

Position Summary

The AI Infrastructure Solutions Architect is responsible for designing end-to-end AI infrastructure solutions that enable customers to deploy scalable, high-performance AI and HPC environments. The role combines technical presales, solution architecture, infrastructure sizing, and deployment planning to deliver commercially viable and operationally sound AI infrastructure solutions.

The architect serves as the technical authority for compute, networking, storage, AI software stack, and infrastructure integration while working closely with Sales, Delivery, Product Management, OEM partners, and the Facilities Presales & Design teams to ensure that the AI infrastructure requirements are fully aligned with the data center's power, cooling, space, and physical infrastructure capabilities.

This role owns the AI infrastructure architecture and complements the facility design process.

What you will do

Key Responsibilities

AI Infrastructure Presales

Partner with Sales to qualify AI infrastructure opportunities.

Engage with customers to understand AI workloads, performance expectations, scalability requirements, and operational objectives.

Conduct technical discovery workshops and architecture discussions.

Prepare technical proposals, presentations, and solution demonstrations.

Support RFI, RFP, and RFQ responses.

Serve as the trusted technical advisor throughout the sales lifecycle.

AI Infrastructure Solution Architecture

Design complete AI infrastructure solutions including:

GPU compute clusters

AI Factory infrastructure

HPC platforms

High-performance storage

AI networking fabrics (Ethernet, InfiniBand, RoCE)

Kubernetes and container platforms

AI software stack integration

Cluster management

Security architecture

Monitoring and observability

Automation platforms

Scalability and expansion planning

Develop:

Solution architecture

Infrastructure sizing

Bill of Materials (BoM)

Reference architectures

Technical specifications

Solution documentation

Deployment architecture

Infrastructure Integration with Data Center Design

Work closely with Facilities Presales, Design, and Engineering teams to ensure the AI infrastructure solution is fully compatible with the proposed data center environment.

Provide technical inputs related to:

Rack power density

Rack layouts

GPU server deployment

Network topology

Cable density

Space requirements

Liquid cooling interfaces

CDU connectivity requirements

Infrastructure dependencies

Expansion strategy

Validate that the AI infrastructure can be successfully deployed within the facility constraints without owning the facility design itself.

Solution Validation & Delivery Readiness

Collaborate with Delivery and Engineering teams to ensure solution feasibility.

Support:

Technical design reviews

OEM interoperability validation

Deployment planning

Installation readiness

Factory acceptance planning

Site acceptance planning

Commissioning support

Technical handover to delivery teams

Maintain ownership of the AI infrastructure solution throughout the project lifecycle.

Operational Architecture

Ensure the proposed solution is designed for long-term operational success by considering:

Serviceability

Scalability

High availability

Redundancy

Cluster management

Monitoring

Capacity management

Firmware lifecycle

Upgrade strategy

Infrastructure observability

Operational support requirements

Cross-Functional Collaboration

Work closely with:

Sales

Product Management

Delivery

Professional Services

Facilities Presales

Facilities Design & Engineering

OEM partners

System Integrators

Customer Infrastructure teams

AI Engineering teams

Act as the primary technical interface for all AI infrastructure-related discussions.

Technical Expertise

AI Compute

NVIDIA GPU platforms

AMD GPU platforms

Intel AI platforms

GPU cluster design

AI Factory architectures

HPC infrastructure

Networking

Ethernet (100/200/400/800G)

InfiniBand

RoCE

Spine-Leaf architectures

AI fabric design

Storage

Parallel file systems

High-performance NAS

Object storage

NVMe-over-Fabrics

AI data pipelines

Software

Kubernetes

Docker

Slurm

NVIDIA AI Enterprise

GPU scheduling