Forward Deployed Engineer - IT Infrastructure

About Hosted AI :

Hosted AI is building the orchestration and virtualization layer for the GPU cloud era, enabling NeoClouds, GPU operators, and infrastructure providers to turn raw GPU hardware into monetizable, multi-tenant compute services.

We're early-stage and moving fast: every customer engagement is a proof point, and the Forward Deployed Engineering function is where technical credibility gets built or lost in real time.

The Role :

We're hiring a Forward Deployed Engineer to own the technical backbone of our customer engagements, from first technical discovery call through POC execution to production handoff. You'll work embedded with our customers and side by side with Sales and Account Management to translate a prospect's infrastructure reality into a working, benchmarked Hosted AI deployment.

This is a hands-on, builder's role. You will be in the admin panel, in the cluster configs, in the Kubernetes YAML, and on the call explaining GPU multiplexing to a data center operator.

What You'll Do :Customer Discovery & Solution Design :

• Run technical discovery with customers and prospects to understand their GPU, compute, network, storage, existing virtualization approach, and success criteria.

• Translate business requirements into a concrete architecture: instance/resource policies, image and pricing policies, team/workspace structure, and networking.

• Advise on GPU virtualization strategy: when to use GPU-as-a-Service vs. KVM-based CPU virtualization, and how to structure mixed GPU/CPU Kubernetes node pools.

Deployment & POC Ownership :

• Own the proof-of-concept end to end: environment setup, cluster and node configuration, recipe (Ansible-based) customization, and validation against success criteria.

• Deploy and configure Hosted AI directly in customer environments, and carry each engagement through to production handoff.

• Track POC health against target timelines and proactively flag stalled or at-risk engagements.

• Build and maintain reference configurations aligned to NVIDIA reference architectures (NCP, Enterprise RA) and common inference/training stack patterns (vLLM, SGLang, NVIDIA AI Enterprise).

Customer-Facing Technical Leadership :

• Deliver live product demos, technical presentations, and RFP/RFI responses.

• Handle in-cycle technical objections (multi-tenancy, SLA/fault-tolerance, CUDA/driver compatibility, networking) with direct, technically accurate answers.

• Partner with Sales and Account Management to identify and develop technical champions within each account.

Field Feedback & Enablement :

• Log product gaps, feature requests, and recurring technical objections surfaced in the field, and route them to Product/Engineering.

• Contribute to internal battlecards, reference architecture docs, and POC playbooks.

• Bring cross-deal patterns back to the Director of Solutions Engineering.

What You Bring :

• 8+ years in a forward deployed, pre-sales, solutions engineering, or solutions architecture role at an infrastructure, cloud, or virtualization company (GPU/AI infra strongly preferred).

• Working knowledge of virtualization concepts (KVM or equivalent) and container orchestration (Kubernetes).

• Hands-on Linux systems experience; comfortable reading and modifying Ansible playbooks or similar configuration-as-code tooling.

• Familiarity with GPU hardware classes (H100/H200/L40S/RTX 6000-class) and the practical differences that drive architecture decisions.

• Ability to run a technical discovery call solo.

• Excellent written and verbal communication.

Strongly Preferred :

• Exposure to NVIDIA reference architectures and modern inference/serving stacks (vLLM, SGLang, NVIDIA AI Enterprise).

• Experience selling into or supporting NeoClouds, GPU operators, or emerging cloud/hosting providers.

• Startup or early-stage experience.

• Familiarity with structured sales methodologies (MEDDPICC, Gap Selling).

• Certifications: CKA (Certified Kubernetes Administrator), NCA-AIIO (NVIDIA-Certified Associate: AI Infrastructure and Operations).