Intermediate9h 32m4 modules · 22 leçons0 inscrits

Augmenting an LLM with RAG — the NVIDIA endpoints

Your first RAG that actually works. Across four hands-on modules, you build a complete chain — ingestion, chunking, embeddings, hybrid retrieval, reranking, cited generation, quantified evaluation — on hosted NVIDIA endpoints (build.nvidia.com), with no GPU to buy and no cluster to manage. The course ends with an honestly graded project: recall, faithfulness, p95 latency. It's the bridge between RAG theory (AI Foundations course 06) and enterprise-scale NIM/NeMo deployment (NVIDIA course 06).

Par Équipe Curriculum Nexus

99€

Accès complet · Certificat inclus

S'inscrire maintenant

Programme du cours

4 modules

01
Module
Augmenting an LLM with RAG
Get started with Augmenting an LLM with RAG in the NVIDIA ecosystem.
5 leçons·2h 01

Production-grade ingestion, chunking, and embeddings

RAG quality is decided 70% upstream: how to ingest, chunk, and vectorize correctly.

5 leçons·1h 50

Augmenting an LLM with RAG — the NVIDIA endpoints

Programme du cours

Augmenting an LLM with RAG

Production-grade ingestion, chunking, and embeddings

Advanced retrieval: rerank, hybrid search, query transformation

RAG in production: evaluation, security, enterprise capstone