ADR 0016: Adopt NVIDIA Certification Framework for Community AI Skills Development

Proposed Status: Proposed Date: 2026-05-08 Domain: agents Level: system Authors: Genesis
nvidiacertificationai-skillstraining

Context

RegenNeighbourhood members need a clear AI skill path. The community lacks a structured skill-verification system.

D1: Adopt three Associate certs as baseline

The community adopts three NVIDIA Associate certifications:

  • Generative AI LLM — LLM fundamentals, prompt engineering, RAG, fine-tuning.
  • Agentic AI — multi-agent workflows, autonomous decision systems.
  • AI Infrastructure and Operations — edge node management, GPU orchestration.

D2: Use free content for onboarding

Start with freely available NVIDIA learning paths:

  • OpenUSD Foundations (11h free) — 3D digital twin literacy.
  • Robotics Fundamentals (28h free) — farm automation foundation.
  • NIM Microservices intro (2h free) — inference deployment.
  • Decentralized AI with FLARE (4h free) — federated learning.

D3: Track progress via community skills registry

Members log course completions and badge IDs in a shared registry. Three states: Interest, In Progress, Certified.

Positive

  • Clear vendor-aligned progression framework.
  • Free content tracks lower the barrier for initial engagement.
  • Associate certifications provide verifiable skill markers.
  • Physical AI and digital twin tracks map to RegenNeighbourhood use cases.

Negative

  • NVIDIA certifications need internet for exams.
  • Exam costs (USD 125–400) may limit participation.
  • Hardware-centric framework. May be overkill for pure software work.
  • No community-specific content in standard NVIDIA curriculum.