• Lead technical discovery and solution positioning for enterprise AI infrastructure opportunities, translating business outcomes into reference architectures and value propositions. • Own pre-sales deliverables including architectures, diagrams, sizing, BOMs, proposals. • Deliver executive and technical presentations focused on NVIDIA AI Enterprise (NVAIE), LLM training/inference, and accelerated analytics. • Guide clients through technology selection, roadmap development, and business case creation for large-scale AI initiatives. • Architect end-to-end AI platforms using NVIDIA DGX/HGX, Blackwell (B100/B200), Hopper (H100/H200), Grace/Grace-Hopper (GH200), L40S, NVLink/NVSwitch, InfiniBand (NVIDIA Quantum), RoCE, and DPU offload patterns. • Design solutions leveraging AMD Instinct (MI300/MI300X) as appropriate, articulating trade-offs in CPU/GPU/DPU, interconnect topology, and cluster scale-out. • Integrate NVIDIA AI Enterprise components (CUDA, cuDNN, TensorRT, Triton Inference Server, RAPIDS) and common ML frameworks (PyTorch, TensorFlow) with orchestration platforms. • Experience integrating on-prem GPU clusters with cloud AI services (AWS SageMaker, Azure ML, GCP Vertex AI) for hybrid bursting and workload mobility. • Advise on MLOps platforms (MLflow, Kubeflow, Weights & Biases), CI/CD, and governance for multi-tenant AI environments. • Build and maintain relationships with NVIDIA, AMD, Run:AI, OEMs, and networking vendors, aligning campaigns with partner programs and incentives. • Contribute feedback to vendor engineering and product teams, coordinating joint enablement and reference designs. • Create repeatable assets such as validated designs, sizing calculators, POV guides, deployment runbooks, and competitive playbooks. • Mentor SEs and delivery consultants, leading internal training on AI scheduling, performance tuning, and operational best practices. • Lead proof-of-value (POV) and proof-of-concept (POC) engagements, including success criteria, benchmarking, and recommendations.