← Workshop overview

The details

Building Agentic AI Applications With LLMs

Delivery

Private industry cohorts anywhere in the United States, in person at your offices or online. NVIDIA supplies the cloud labs. Allow six weeks to schedule.

Duration
Eight hours · hands-on
Your setup
NVIDIA cloud GPU labs. No local GPUs needed.

NVIDIA owns the content, cloud GPU labs, assessment, and the DLI certificate. Nexus hosts and teaches with a Certified Instructor, handles enrollment and invoices the client.

Nexus pricing

$500per seat

$500 per seat for groups up to 20, invoiced by Nexus. For 21 or more we send a tailored quote. Up to 40 per cohort for the best hands-on results; larger teams run as multiple cohorts.

Before you join

Check the course prerequisites and system requirements before enrolling.

NVIDIA course requirements

What you’ll cover

Agent fundamentals

What an LLM agent is, where language models are strong, and where they fail, so your team can tell an agent-shaped problem from one that needs ordinary software.

Structured outputs and tool use

Constraining a model to machine-readable output so its responses can drive function calls and API integrations reliably.

Retrieval and knowledge graphs

Grounding agents in your domain knowledge with retrieval mechanisms and graph-backed context.

Multi-agent systems with LangGraph

Decomposing work across specialist agents, giving them communication channels, and running them concurrently with LangGraph orchestration.

Deployment and final assessment

Deploying an agent that schedules multiple retrieval operations and reports back, the hands-on assessment behind the NVIDIA DLI certificate.

NVIDIA NIM · build.nvidia.com · LangChain · LangGraph · Python · PyTorch

Full course and assessment details