NVIDIA Generative AI LLM Professional

Cert track · NVIDIA self-paced

The NVIDIA self-paced courses in this cert prep path. Free courses do not require Inception codes; paid courses redeem one code each.

• S-FX-33Substitute

  Domain-Adaptive Pre-Training: Tailoring LLMs for Specialized Applications

  4h
• S-FX-24Substitute

  The Art of Compressing LLMs: Pruning, Distillation, and Quantization

  8h
• S-FX-18Substitute

  Sizing LLM Inference Systems

  3h
• S-FX-15

  Building RAG Agents with LLMs

  8h
• S-FX-19

  Introduction to Deploying RAG Pipelines for Production at Scale

  4h

Cert track · BakedIn parallel courses

BakedIn cert-prep guides being authored against the same exam blueprint. Validated by Glen passing the cert from the BakedIn content.

• BakedIn LLM Domain Adaptation + Compression

  Pairs to S-FX-33 + S-FX-24 (substitutes for Adding New Knowledge workshop). DAPT, SFT, DPO, quantization, pruning, distillation.

• BakedIn LLM Inference at Scale

  Pairs to S-FX-18 (substitute for Deploying & Optimizing Inference workshop). Inference economics, model serving, latency optimization. Backed by Bedrock-routed per-task model selection.

• BakedIn RAG Foundations

  Shared with NCP-AAI track.

• BakedIn RAG Production

  Shared with NCP-AAI track.

How this prep is built