- Design end-to-end GPU infrastructure architectures for enterprise, regulated, and sovereign AI deployments
- Create GPU sizing calculators and tiered reference architectures for H100/H200/B200-class deployments
- Lead cluster validation, benchmark planning, readiness assessments, and performance tuning
- Advise on air-gapped and private deployment patterns including VPC isolation, registry mirroring, and on-prem NIM hosting
- Engage NVIDIA, hyperscaler, and OEM stakeholders to finalize architecture decisions and resolve blockers
- Develop whitepapers, reference architecture decks, TCO models, and enablement assets