Director, Presales Solution Architecture – NeoCloud

This is a fully remote position, open to applicants in Kazakhstan, +1 more country.

📋 Description

• Lead and develop the sales engineering and solutions architecture team.

• Recruit, mentor, and retain sales engineers and solutions architects.

• Establish the pre-sales operating model as the organization grows.

• Create discovery frameworks, reference architectures, TCO/benchmark models, POV playbooks, demo and benchmark environments, and RFP response libraries.

• Set and uphold a technical quality standard across the team.

• Conduct enablement sessions on GPU architecture, networking, and orchestration.

• Collaborate with Account Executives as the technical lead on strategic and enterprise opportunities from discovery to technical close.

• Execute qualification using MEDDPICC or a similar methodology.

• Identify the economic buyer, decision criteria, and technical champion, and develop win plans.

• Design solutions across compute, networking, storage, and orchestration.

• Generate sizing, capacity plans, and TCO comparisons against hyperscalers and self-built alternatives.

• Design and drive POCs/POVs, define success criteria, conduct benchmarks, and translate results into commercial momentum.

• Provide product and capacity requirements to product, platform, and supply/capacity planning teams.

• Collaborate with NVIDIA field and partner ecosystem on joint pursuits and reference architectures.

• Influence product roadmap and packaging based on field inputs.

⛳️ Requirements

• Hands-on experience with AI/ML infrastructure, including running or establishing distributed training and/or production inference workloads.

• Practical knowledge of data pipelines, distributed training (multi-node/multi-GPU), fine-tuning, and serving.

• Understanding of interconnect, memory bandwidth, I/O, and scheduling bottlenecks.

• Experience with PyTorch and related tools, including NCCL, CUDA-level concepts, containers, and schedulers.

• In-depth knowledge of NVIDIA compute platforms, including Hopper and Blackwell generations, H100/H200, GB200 NVL72, B200-class systems, Grace-Hopper, DGX, HGX, and MGX.

• Familiarity with NVLink/NVSwitch, InfiniBand, Spectrum-X Ethernet, RDMA/RoCE, and DPUs.

• Understanding of NVIDIA AI Enterprise, NIM, NeMo, Triton/TensorRT-LLM, Base Command, Run:ai/GPU orchestration, and NGC.

• Knowledge of NVIDIA Cloud Partner motion and co-selling with NVIDIA.

• Proven track record in supporting complex B2B deals with sales cycles of 6–18+ months and significant ACV/TCV.

• Experience with multi-year committed-capacity or reserved-capacity structures is advantageous.

• Proficient in engaging with ML/infra, platform engineering, procurement, finance, security, and executive stakeholders.

• Ability to construct and defend TCO/ROI models against hyperscaler and on-premises options.

• Capable of translating performance benchmarks into commercial value.

• Experience in hiring, developing, and leading a sales engineering/solutions architecture team, or demonstrable readiness to do so.

• Player-coach mindset.

• Strongly preferred: experience in selling GPU cloud, HPC, or specialized infrastructure.

• Strongly preferred: knowledge of Kubernetes, GPU operators/device plugins, Slurm, multi-cluster management, and storage-for-AI.

• Familiarity with virtualized GPU/KubeVirt-style patterns is a plus.

• Experience with data center economics and limitations, including power, cooling, rack density, and capacity availability.

• Exposure to sovereign, regulated, or government AI buyers.

🏝️ Benefits

• Opportunities for professional development and training.

• Participation in conferences and working groups.

• Company outings, happy hours, hackathons, and tech talks.

• Competitive compensation package accompanied by a robust benefits plan.

• Flexible remote work arrangement (employees have the option to work remotely).