Personas

Solutions apply an IT Admin + functional SME + AI engineer lens to NVIDIA's developer-focused reference materials. Every solution states how it serves each persona.

IT Admin

• Cares about: system setup, drivers/OS, containerized deployment, fleet management, identity (AD/LDAP), monitoring, scaling dev→prod, security.
• Solution gives them: step-by-step setup, management-environment config, and fleet/scaling guidance.

Functional SME

• Cares about: does this use case fit my workflow? what scale/quality should I expect? what does "good" look like?
• Solution gives them: the use-case framing, S/M/L scale guidance (user counts, expected performance), and validation results.

AI Engineer

• Cares about: model choice, quantization, inference stack tuning, context length, throughput/latency, integration (OpenAI-compatible API, tools).
• Solution gives them: per-tier / per-workload model recommendations (Nemotron-family primary + measured alternatives — the stack guide is the source of truth for tier-specific winners), vLLM configuration, reference code/config, and aiperf benchmarks.

Mapping to solution sections