Nvidia-Learning-Training Course-Catalog
Nvidia-Learning-Training Course-Catalog
With access to high-performance computing, you’ll learn how to train, optimize, and deploy
neural networks using the latest deep learning tools, frameworks, and SDKs. You’ll also learn
how to assess, parallelize, optimize, and deploy NVIDIA - accelerated computing applications.
Our training program offers both self-paced online courses and instructor-led, prescheduled
workshops. The self-paced courses range from ten minutes to eight hours and guide you
through applying a specific technology, setting up a project, or administering solutions in a
data center. Instructor-led workshops and boot camps go deeper into topic areas, teaching
you how to implement a project or solution from end to end. Both types of courses give you
valuable hands-on experience using the latest technologies.
> Gain hands-on experience with the most widely used, industry-standard platforms including
software, hardware, tools, and frameworks. Each student will have access to a fully configured,
GPU-accelerated server in the cloud or access to NVIDIA solutions in our training lab.
> Become proficient in administering NVIDIA hardware and software solutions such as DGX™,
InfiniBand, Cumulus, NVIDIA AI Enterprise, and more.
> Access instructor-led workshops and online courses from anywhere using just a laptop and
internet connection.
> Acquire real-world expertise through content designed in collaboration with industry leaders
such as Children’s Hospital of Los Angeles, Mayo Clinic, and PwC.
> Earn NVIDIA certifications and course completion certificates to indicate subject matter
competency and support your career growth.
Data Science
Accelerating Data Engineering Pipelines 8
Enhancing Data Science Outcomes With Efficient Workflows 8
Fundamentals of Accelerated Data Science 8
Deep Learning
Applications of AI for Anomaly Detection 9
Applications of AI for Predictive Maintenance 9
Building AI-Based Cybersecurity Pipelines 9
Building Conversational AI Applications V2.0 10
Building Deep Learning-Based Anti-Fraud Applications (Chinese only) 10
Building Transformer-Based Natural Language Processing 10
Computer Vision for Industrial Inspection 10
Data Parallelism: How to Train Deep Learning Models on Multiple GPUs 11
Getting Started With AI on NVIDIA® Jetson Nano 11
™
Data Science
Accelerate Data Science Workflows With Zero Code Changes 16
Accelerating End-to-End Data Science Workflows 16
RAPIDS Accelerator for Apache Spark 16
Deep Learning
Building a Brain in 10 Minutes 16
Building Real-Time Video AI Applications 16
Deploying a Model for Inference at Production Scale 17
Digital Fingerprinting With Morpheus 17
Disaster Risk Monitoring Using Satellite Imagery 17
Exploring Adversarial Machine Learning 17
Get Started With Highly Accurate, Customer ASR for Speech AI 18
Getting Started With AI on Jetson Nano 18
Getting Started With Deep Learning 18
Getting Started With Image Segmentation 18
Integrating Sensors With NVIDIA DRIVE 18
Introduction to Graph Neural Networks 19
Introduction to Physics-Informed Machine Learning With NVIDIA Modulus 19
Infrastructure
AI Infrastructure and Operations Fundamentals 26
Cluster Administration
NVIDIA Base Command™ Manager 27
Ethernet Cumulus
Cumulus® Linux: Public Bootcamp 27
Cumulus Linux: Private Workshop 27
NVIDIA Cumulus Linux: Customized Advanced Training 27
InfiniBand
InfiniBand Customized Course 28
InfiniBand Professional Customized Training 28
NVIDIA DGX
NVIDIA DGX H200/H100/A100 Administration: Private Workshop 28
NVIDIA DGX H200/H100/A100 Administration: Public Workshop 28
NVIDIA DGX BasePOD Administration: Private Workshop
™
29
NVIDIA DGX SuperPOD Administration: Private Workshop
™
29
Virtualization
NVIDIA AI Enterprise Administration: Public Bootcamp 29
Cluster Administration
NVIDIA Base Command™ Manager 30
Base Command Manager Autoscaling Hybrid Cloud 31
Introduction to Base Command Manager 31
Ethernet
Network Administration With the NVIDIA Onyx™ Switch System 31
RDMA Over Converged Ethernet (RoCE) From A to Z 31
InfiniBand
InfiniBand Essentials 31
InfiniBand Professional 32
Management
Data Center Management Made Easy With NVIDIA UFM® 32
NVIDIA License System 32
Network
Ansible Essentials for Network Engineers 32
Introduction to Networking 33
MLXlink and MLXcables Debug Tools 33
NVIDIA BlueField® DPU Administration 33
RDMA
The Fundamentals of RDMA Programming 33
Certifications
NVIDIA-Certified Associate: AI Infrastructure and Operations 34
NVIDIA-Certified Associate: Generative AI Large Language Models 34
NVIDIA-Certified Associate: Generative AI Multimodal 34
NVIDIA-Certified Professional: InfiniBand 34
Instructor-Led Workshops for Developers | Accelerated Computing
Accelerated Computing
Accelerating CUDA® Discover how to write CUDA C++ applications Professional experience programming CUDA C/
C++ Applications With that efficiently and correctly use all available C++ applications, including the use of the NVIDIA
Multiple GPUs GPUs in a single node, dramatically improving the CUDA Compiler (NVCC), kernel launches, grid-
performance of applications and making the most stride loops, host-to-device and device-to-host
cost-effective use of systems with multiple GPUs. memory transfers, and CUDA error handling.
Familiarity with the Linux command line and
> Learn More
experience using makefiles to compile C/C++ code.
Fundamentals of Learn how to accelerate and optimize existing C/ Basic C/C++ competency, including familiarity
Accelerated Computing C++ CPU-only applications to apply the power of with variable types, loops, conditional statements,
With CUDA C/C++ GPUs using the most essential CUDA techniques functions, and array manipulations. No previous
and the NVIDIA Nsight Systems profiler. knowledge of CUDA programming is assumed.
> Learn More
Fundamentals of Explore how to use Numba—the just-in-time, type- Basic Python competency, including familiarity
Accelerated Computing specializing Python function compiler—to create with variable types, loops, conditional statements,
With CUDA Python and launch CUDA kernels to accelerate Python functions, and array manipulations. Also, must
programs on massively parallel NVIDIA GPUs. have NumPy competency, including the use of
ndarrays and ufuncs.
> Learn More
Fundamentals of Find out how to write and configure code Basic C/C++ or Fortran competency, including
Accelerated Computing parallelization with OpenACC, optimize memory familiarity with variable types, loops, conditional
With OpenACC® movements between the CPU and GPU accelerator, statements, functions, and array manipulations.
and apply the techniques to accelerate a CPU-only No previous knowledge of GPU programming is
Laplace heat equation to achieve performance gains. assumed.
> Learn More
Back
7 | OCT24 | NVIDIA Training Course Catalog
Instructor-Led Workshops for Developers | Accelerated Computing
Scaling CUDA C++ Learn the tools and techniques needed to write Intermediate experience writing CUDA C/C++
Applications to CUDA C++ applications that can scale efficiently applications.
Multiple Nodes to clusters of NVIDIA GPUs.
> Learn More
Data Science
Accelerating Data Explore how to employ advanced data engineering Intermediate knowledge of Python (list
Engineering Pipelines tools and techniques with GPUs to significantly comprehension, objects). Familiarity with pandas
improve data engineering pipelines. and introductory statistics (mean, median,
mode) a plus.
> Learn More
Enhancing Data Learn how to create an end-to-end, hardware- > Basic knowledge of a standard data
Science Outcomes accelerated machine learning pipeline for large science workflow on tabular data.
With Efficient datasets. Throughout the development process, > Knowledge of distributed computing
Workflows you’ll use diagnostic tools to identify delays and using Dask.
learn to mitigate common pitfalls. > Completion of the DLI’s Fundamentals of
> Learn More Accelerated Data Science course or an
ability to manipulate data using cuDF and
some experience building machine learning
models using cuML.
Fundamentals of Learn how to perform multiple analysis tasks on Professional data science experience with
Accelerated Data large datasets using NVIDIA RAPIDS™, a collection Python, including proficiency in pandas and
Science of data science libraries that allows end-to-end GPU NumPy. Also, must have familiarity with common
acceleration for data science workflows. machine learning algorithms, including XGBoost,
linear regression, DBSCAN, K-Means, and SSSP.
> Learn More
Back
8 | OCT24 | NVIDIA Training Course Catalog
Instructor-Led Workshops for Developers | Deep Learning
Deep Learning
Applications of AI for Learn to detect anomalies in large datasets to Experience with convolutional neural networks
Anomaly Detection identify network intrusions using supervised and (CNNs) and Python.
unsupervised machine learning techniques, such as
accelerated XGBoost, autoencoders, and generative
adversarial networks (GANs).
> Learn More
Applications of AI for Discover how to identify anomalies and failures in Experience with Python and deep networks.
Predictive Maintenance time-series data, estimate the remaining useful
life of the corresponding parts, and use this
information to map anomalies to failure conditions.
> Learn More
Building AI-Based Traditional cybersecurity methods include creating > Professional data science and/or data analysis
Cybersecurity Pipelines barriers around your infrastructure to protect it experience.
from intruders. However, as enterprises continue to > Competency with the Python programming
digitally transform, they’re faced with a proliferation language.
of devices, more sophisticated cybersecurity > Competency with the Linux command line.
attacks, and an incredibly vast network of data
to protect—which means new cybersecurity
methodologies must be explored. An alternative
approach is to address cybersecurity as a data
science problem: Better understand all the users
and activities across your network so that you can
identify which transactions are typical and which
are potentially nefarious.
The NVIDIA Morpheus AI framework lets
cybersecurity developers and practitioners
harness the power of GPU computing to
implement cybersecurity solutions that perform
on a scale never before possible. With Morpheus,
cybersecurity developers can create optimized AI
pipelines for filtering, processing, and classifying
large volumes of real-time data. Bringing a new
level of information security to data centers,
Morpheus enables dynamic protection, real-time
telemetry, and adaptive defenses for detecting and
remediating cybersecurity threats.
> Learn More
Back
9 | OCT24 | NVIDIA Training Course Catalog
Instructor-Led Workshops for Developers | Deep Learning
Building Conversational Discover how to quickly build and deploy Experience with Python coding and use of library
AI Applications V2.0 production-quality speech AI applications with real- functions and parameters. Also, a fundamental
time transcription and natural language processing understanding of a deep learning framework,
capabilities. such as TensorFlow, PyTorch, or Keras, and a
basic understanding of neural networks.
> Learn More
Building Deep This course is primarily for data scientists and > Basic Python programming experience.
Learning-Based Anti- professionals working in the field of financial > Fundamental understanding of deep learning
Fraud Applications fraud modeling in banks. It teaches how to train, frameworks (such as TensorFlow, PyTorch,
(Chinese only) accelerate, and optimize fraud detection classifiers or Keras).
based on machine learning and deep learning. > Basic knowledge of neural networks.
> Learn More
Building Transformer- In this workshop, you’ll learn how Transformers are Experience with Python coding and use of
Based Natural used as the building blocks of modern large language library functions and parameters. Fundamental
Language Processing models (LLMs). You’ll then use these models for understanding of a deep learning framework,
various NLP tasks, including text classification, such as TensorFlow, PyTorch, or Keras. And basic
named-entity recognition (NER), author attribution, understanding of neural networks.
and question answering. You’ll also learn how to
analyze various model features, constraints, and
characteristics to determine which model is best
suited for a particular use case based on metrics,
domain specificity, and available resources.
> Learn More
Computer Vision for In this workshop, you’ll learn how to quickly develop > Experience with Python; basic understanding
Industrial Inspection and deploy a machine learning model that uses of data processing and deep learning
deep learning for computer vision to perform defect
classification and other visual recognition tasks. > To gain experience with Python, we suggest
Using NVIDIA’s own real production dataset as an this Python tutorial
example, this workshop illustrates how the solution > For a basic understanding of data processing
can be easily applied to a variety of manufacturing and deep learning, we suggest Fundamentals
and industrial inspection use cases. of Deep Learning.
> Learn More
Back
10 | OCT24 | NVIDIA Training Course Catalog
Instructor-Led Workshops for Developers | Deep Learning
Data Parallelism: This workshop teaches you techniques for data- Experience with deep learning training using
How to Train Deep parallel deep learning training on multiple GPUs Python. See the Fundamentals of Deep Learning
Learning Models on to shorten the training time required for data- self-paced course here.
Multiple GPUs intensive applications. Working with deep learning
tools, frameworks, and workflows to perform neural
network training, you’ll learn how to decrease model
training time by distributing data to multiple GPUs,
while retaining the accuracy of training on a
single GPU.
> Learn More
Getting Started Build and train a classification dataset and model Basic familiarity with Python
With AI on NVIDIA® with NVIDIA Jetson Nano.
Jetson Nano™
> Learn More
Fundamentals of Deep Learn how deep learning works through hands-on An understanding of fundamental programming
Learning exercises in computer vision and natural language concepts in Python 3, such as functions, loops,
processing (NLP). You’ll train deep learning models dictionaries, and arrays. Also, familiarity with
from scratch and pick up tricks and tools for pandas data structures and an understanding of
achieving highly accurate results along the way. how to compute a regression line.
You’ll also learn to leverage freely available, state-of- > Suggested materials to satisfy prerequisites:
the-art pretrained models to save time and get your Python Beginner’s Guide
deep learning application up and running quickly.
> Learn More
Model Parallelism: In this workshop, you’ll learn how to scale training > Good understanding of PyTorch, deep learning,
Building and Deploying and deployment of LLMs and neural networks and data parallel training concepts
Large Neural Networks across multiple nodes, use various forms of model > Practice with multi-GPU training and natural
parallelism to overcome the challenges associated language processing is useful, but optional.
with large-model memory footprint, capture and
understand training performance characteristics to
optimize model architecture and deploy very large
multi-GPU, multi-node models to production using
NVIDIA Triton™ Inference Server.
> Learn More
Back
11 | OCT24 | NVIDIA Training Course Catalog
Instructor-Led Workshops for Developers | Generative AI and Large Language Models
Building Transformer- Learn how to apply and fine-tune a transformer- > Experience with Python coding and use of
Based Natural based deep learning model to natural language library functions and parameters.
Language Processing processing (NLP) tasks. > Fundamental understanding of a deep learning
Application framework such as TensorFlow, PyTorch, or
> Learn More
Keras.
> Basic understanding of neural networks.
Building LLM In this course, you’ll go beyond prompt engineering This course is primarily intended for
Applications With LLMs and learn a variety of techniques to efficiently intermediate level and above Python developers
Prompt Engineering customize pretrained LLMs for your specific use with a solid understanding of LLM fundamentals
cases—without engaging in the computationally and some prompt engineering experience.
intensive and expensive process of pretraining
your own model or fine-tuning a model’s internal
weights. Using NVIDIA NeMo™ service, you’ll learn
various parameter-efficient fine-tuning methods to
customize LLM behavior for your organization.
> Learn More
Efficient Large Learn a variety of techniques to efficiently > Professional experience with the Python
Language Model customize pretrained LLMs for your specific use programming language.
Customizations cases—without engaging in the computationally > Familiarity with fundamental deep learning
intensive and expensive process of pretraining topics like model architecture, training and
your own model or fine-tuning a model’s internal inference.
weights. Using the open-source NVIDIA NeMo > Familiarity with a modern Python-based deep
framework, you’ll learn prompt engineering and learning framework (PyTorch preferred).
various parameter-efficient fine-tuning methods to
> Familiarity working with out-of-the-box
customize LLM behavior for your organization.
pretrained LLMs.
> Learn More
Back
12 | OCT24 | NVIDIA Training Course Catalog
Instructor-Led Workshops for Developers | Generative AI and Large Language Models
Generative AI With Get started with gen AI application development > Good understanding of PyTorch
Diffusion Models with this hands-on course where you’ll learn how > Good understanding of deep learning
to build a text-to-image generative AI application
using the latest techniques. Generate images with
diffusion models and refine the output with various
optimizations. Build a denoising diffusion model
from the U-Net architecture to context embeddings
for greater user control.
> Learn More
Rapid Application In this course, you’ll gain a strong understanding > Introductory deep learning, with comfort
Development Using and practical knowledge of LLM application with PyTorch and transfer learning preferred.
Large Language Models development by exploring the open-source Content covered by DLI’s Getting Started
ecosystem, including pretrained LLMs, that can With Deep Learning or Fundamentals of Deep
help you get started quickly developing LLM-based Learning courses, or similar experience is
applications. sufficient.
> Learn More > Intermediate Python experience, including
object-oriented programming and libraries.
Content covered by Python Tutorial (w3schools.
com) or similar experience is sufficient.
Bootstrapping Learn how to use NVIDIA Omniverse Replicator, > Intermediate understanding of Python
Computer Vision a core Omniverse extension, to accelerate the (including classes, objects, and decorators).
Models with Synthetic development of computer vision models. Generate > Basic understanding of Machine Learning and
Data accurate, photorealistic, physics-conforming Deep Learning concepts and pipelines.
synthetic data to ease the expensive, time-
consuming task of labeling real-world data.
Omniverse Replicator accelerates AI development at
scale and reduces time to production.
> Learn More
Building Digital Avatar This course, from an end-to-end application > Basic Python programming experience.
Pipelines With NVIDIA development perspective, will provide you with > Fundamental understanding of deep
Omniverse Audio2Face detailed guidance on how to use NVIDIA Omniverse neural networks.
and Riva (Chinese only) Audio2Face and the interactive speech suite Riva to
build virtual digital humans.
> Learn More
Back
13 | OCT24 | NVIDIA Training Course Catalog
Online, Self-Paced Courses for Developers | Accelerated Computing Fundamentals
Accelerating CUDA Discover how to improve performance for your Professional experience programming CUDA C/
C++ Applications With CUDA C/C++ applications by overlapping memory C++ applications, including the use of the nvcc
Concurrent Streams transfers to and from the GPU with computations compiler, kernel launches, grid-stride loops,
on the GPU. host-to-device and device-to-host memory
transfers, and CUDA error handling; Experience
> Learn More
using Makefiles to compile C/C++ code.
An Even Easier Learn the basics of writing parallel CUDA kernels to Competency writing applications in CUDA C/C++.
Introduction to CUDA run on NVIDIA GPUs.
> Learn More
Fundamentals of Explore how to use Numba—the just-in-time, type- Basic Python competency, including familiarity
Accelerated Computing specializing Python function compiler—to create with variable types, loops, conditional
With CUDA Python and launch CUDA kernels to accelerate Python statements, functions, and array manipulations.
programs on massively parallel NVIDIA GPUs. Also, must have NumPy competency, including
the use of ndarrays and ufuncs.
> Learn More
Fundamentals of Find out how to build and optimize accelerated Basic experience with C/C++
Accelerated Computing heterogeneous applications on multiple GPU
With OpenACC clusters using a combination of OpenACC, CUDA-
aware MPI, and NVIDIA profiling tools.
> Learn More
Getting Started With Discover how to accelerate and optimize existing C/ Basic C/C++ competency, including familiarity
Accelerated Computing C++ CPU-only applications to leverage the power with variable types, loops, conditional
With CUDA C/C++ of GPUs using the most essential CUDA techniques statements, functions, and array manipulations.
and the Nsight Systems profiler. No previous knowledge of CUDA programming is
assumed.
> Learn More
Back
14 | OCT24 | NVIDIA Training Course Catalog
Online, Self-Paced Courses for Developers | Accelerated Computing Fundamentals
GPU Acceleration With Learn to write simple, portable, parallel-first Beginner-level experience with C++11. Comfort
the C++ Standard applications using only standard C++ language working with C++ lambdas and standard library
Library features that can be compiled without modification algorithms.
to take advantage of NVIDIA GPU-accelerated
environments.
> Learn More
Optimizing CUDA NVIDIA Developer Tools are a collection of Familiarity with machine learning applications
Machine Learning applications, spanning desktop and mobile targets, using CUDA. We suggest Fundamentals of
Codes With NVIDIA that enable developers to build, debug, profile, and Accelerated Computing with CUDA C/C++.
Nsight™ Profiling Tools develop class-leading and cutting-edge software
using the latest visual computing hardware from
NVIDIA. In this course, you’ll learn the effective use
of two powerful NVIDIA developer tools: Nsight
Systems and Nsight Compute.
Nsight Systems provide developers with a system-
wide visualization of an application’s performance.
Developers can optimize bottlenecks to scale
efficiently across any number or size of CPU and
GPU—from large servers to the smallest systems
on chip. Nsight Compute is an interactive kernel
profiler for CUDA applications. It provides detailed
performance metrics and API debugging via a user
interface and command-line tool.
By the time you complete this course, you’ll be able
to use Nsight Systems and Nsight Compute to
analyze and optimize CUDA applications. Following
best practices, you’ll begin by using Nsight Systems
to analyze overall application structure and explore
parallelization opportunities before turning to
Nsight Compute to analyze and optimize individual
CUDA kernels.
> Learn More
Scaling GPU- In this interactive, hands-on workshop, which is Beginner-level experience with C++11; comfort
Accelerated the followup to GPU Acceleration With the C++ working with C++ lambdas and standard library
Applications With the Standard Library, you’ll learn how to write scalable, algorithms; experience developing C++/MPI
C++ Standard Library GPU-accelerated, hybrid applications using C++ hybrid applications that require inter-rank
standard language features alongside MPI. communication; comfort working with C++
concurrency primitives such as std::thread,
> Learn More
std::barrier, and andstd::thread.
Scaling Workloads Learn how to build robust and efficient CUDA C++ Competency writing applications in CUDA C/C++.
Across Multiple GPUs applications that can take advantage of all available
With CUDA C++ GPUs on a single node.
> Learn More
Back
15 | OCT24 | NVIDIA Training Course Catalog
Online, Self-Paced Courses for Developers | Data Science
Data Science
Accelerate Data In this workshop, you’ll learn to use RAPIDS to Basic understanding of data processing and
Science Workflows speed up your CPU-based data science workflows. knowledge of a standard data science workflow
With Zero Code on tabular data.
> Learn More
Changes
Experience using common Python libraries for
data analytics.
Accelerating End- Explore how to perform multiple analysis tasks Experience with Python, ideally including pandas
to-End Data Science on large datasets using RAPIDS, a collection of and NumPy.
Workflows data science libraries that allows end-to-end GPU
acceleration for data science workflows.
> Learn More
RAPIDS Accelerator In this training lab, we’ll walk through the RAPIDS > Basic experience with Linux terminal
for Apache Spark Accelerator for Apache Spark, including running commands.
SQL queries on CPU and GPU in Spark and diving > Basic experience with Python
into the toolset that helps enable success. > Basic experience with Spark, PySpark,
> Learn More or pandas
Deep Learning
Building a Brain in This one-click notebook explores the biological and An understanding of fundamental programming
10 Minutes psychological inspirations for the world’s concepts in Python 3 such as functions, loops,
first neural networks. dictionaries, and arrays.
> Learn More
Building Real-Time Gain the knowledge and skills needed to enable the Competency in the Python 3, programming
Video AI Applications real-time transformation of raw video data from language, some experience manipulating data
widely deployed camera sensors into deep learning- using pandas DataFrames, and familiarity with
based insights. deep networks (specifically variations of CNNs).
> Learn More
Back
16 | OCT24 | NVIDIA Training Course Catalog
Online, Self-Paced Courses for Developers | Deep Learning
Deploying a Model Learn how to deploy your own machine learning Familiarity with at least one machine learning
for Inference at models on a GPU server. framework, such as PyTorch, TensorFlow, ONNX,
Production Scale or TensorRT.
> Learn More
Digital Fingerprinting In this course, you’ll get hands-on experience This tutorial doesn’t have any prerequisites, but
With Morpheus developing and deploying the NVIDIA digital familiarity with defensive cybersecurity themes
fingerprinting AI workflow that enables 100% and the Linux command line are a plus.
data visibility and drastically reduces the time to
detect threats. You’ll also hear from cybersecurity
experts from a variety of institutions about how to
use NVIDIA AI frameworks and tools to architect
cybersecurity solutions.
> Learn More
Disaster Risk Learn how to build and deploy a deep learning > Competency in the Python 3 programming
Monitoring Using model to automate the detection of flood events language.
Satellite Imagery using satellite imagery. This workflow can be > Basic understanding of machine learning and
applied to lower the cost, improve the efficiency, deep learning concepts, specifically variations
and significantly enhance the effectiveness of of convolutional neural networks (CNNs), and
various natural disaster management use cases. pipelines.
> Interest in understanding how to manipulate
> Learn More
satellite imagery using modern methods.
Exploring Adversarial In this course, which is designed for both data Intermediate experience with Python, ideally
Machine Learning scientists and security practitioners, you’ll explore including PyTorch, pandas, and NumPy.
the security risks and vulnerabilities that adopting
Solid understanding of machine learning and deep
machine learning might expose you to. You’ll also
learning concepts and technologies, as provided in
explore the latest techniques and tools being used
DLI’s Fundamentals of Deep Learning course.
by attackers and build some of your own attacks.
> Learn More These prerequisites can be fulfilled with the
self-paced Getting Started With Deep Learning
course from DLI or other introductory Python-
based machine learning training.
Back
17 | OCT24 | NVIDIA Training Course Catalog
Online, Self-Paced Courses for Developers | Deep Learning
Get Started With Learn to build, train, fine-tune, and deploy a GPU- Basic understanding of machine learning and
Highly Accurate, accelerated automatic speech recognition service deep learning concepts and pipelines.
Customer ASR for with NVIDIA Riva that includes customized features.
In addition, this lab requires that the user have
Speech AI
> Learn More an NVIDIA NGC account and API key.
To fulfill this requirement, please:
> Register and activate a free NGC account
> Generate your NGC API key and save it in a safe
location
Getting Started With Discover how to build a deep learning classification Basic familiarity with Python (helpful, not
AI on Jetson Nano project with computer vision models using the required).
NVIDIA Jetson Nano Developer Kit.
> Learn More
Getting Started With Explore the fundamentals of deep learning by > An understanding of fundamental
Deep Learning training neural networks and using results to programming concepts in Python 3, such as
improve performance and capabilities. functions, loops, dictionaries, and arrays.
> Learn More > Familiarity with pandas data structures
and an understanding of how to compute a
regression line
> Suggested materials to satisfy prerequisites:
Python Beginner’s Guide
Getting Started With Learn how to categorize segments of an image. Basic experience training neural networks.
Image Segmentation
> Learn More
Integrating Sensors Find out how to integrate automotive sensors into Basic experience in C++ and Linux terminal
With NVIDIA DRIVE your applications using NVIDIA DRIVE. commands.
> Learn More
Back
18 | OCT24 | NVIDIA Training Course Catalog
Online, Self-Paced Courses for Developers | Deep Learning
Introduction to Graph Learn the basic concepts, models, and applications Competency in the Python 3 programming
Neural Networks of graph neural networks. language. Experience with deep neural networks
(specifically variations of CNNs).
> Learn More
Introduction to High-fidelity simulations in science and engineering > Familiarity with the Python programming
Physics-Informed are computationally expensive and time-prohibitive language
Machine Learning With for quick iterative use cases, from design analysis to > An understanding of partial differential
NVIDIA Modulus optimization. NVIDIA Modulus, the physics machine equations and their use in physics.
learning platform, turbocharges such use cases by > Familiarity with machine learning concepts like
building physics-based deep learning models that are training and inference.
100,000X faster than traditional methods and offer
high-fidelity simulation results.
Upon completion, you’ll understand the various
building blocks of Modulus and the basics of
physics-informed deep learning. You’ll also
understand how the Modulus framework integrates
with the overall Omniverse platform.
> Learn More
Building RAG Agents Agents powered by LLMs are quickly gaining > Introductory deep learning knowledge, with
for LLMs popularity. An especially powerful recent comfort with PyTorch and transfer learning
development has been the popularization of preferred.
retrieval-based LLM systems that can hold > Intermediate Python experience, including
informed conversations by using tools, looking at object-oriented programming and libraries.
documents, and planning their approaches. This
course will observe how you can deploy an agent
system in practice and scale up your system to
meet the demands of users and customers.
> Learn More
Back
19 | OCT24 | NVIDIA Training Course Catalog
Online, Self-Paced Courses for Developers | Generative AI and Large Language Models (LLMs)
Deploying RAG This course focuses on teaching production- > Familiarity working with LLM-based
Pipelines for level deployment of LLM applications, especially applications
Production at Scale enterprise-grade deployment of RAG pipelines. > Familiarity with RAG pipelines
It covers various aspects for an end-to-end > Familiarity working with Kubernetes
deployment using Helm charts and NVIDIA NIM™
> Familiarity working with Helm
microservices.
> Learn More
Generative AI Generative AI describes technologies that are Basic understanding of machine learning and
Explained used to generate new content based on a variety deep learning concepts
of inputs. In this course, you’ll learn generative AI
concepts, applications, as well as the challenges
and opportunities in this exciting field.
> Learn More
Generative AI With In this workshop, you’ll train deep learning models An understanding of fundamental programming
Diffusion Models from scratch and learn tools and tricks to achieve concepts in Python such as functions, loops,
highly accurate results. You’ll also learn to leverage dictionaries, and arrays.
freely available, state-of-the-art pretrained models
to save time and get your deep learning application
up and running quickly.
> Learn More
Introduction to In this course, you’ll learn how transformers > Basic understanding of deep learning
Transformer-Based are used as the building blocks of modern large concepts.
Natural Language language models (LLMs). You’ll then use these
> Basic understanding of language modeling and
Processing models for various NLP tasks, including text
classification, named-entity recognition (NER), transformers.
author attribution, and question answering.
> Learn More
Prompt Engineering In this course, you’ll interact with and prompt Experience with deep learning training
With Llama 2 engineer Llama 2 models to analyze documents, using Python.
generate text, and be an AI assistant.
> Learn More
Back
20 | OCT24 | NVIDIA Training Course Catalog
Online, Self-Paced Courses for Developers | Generative AI and Large Language Models (LLMs)
Synthetic Tabular Synthetic data generation (SDG) is a data- > Competency in the Python 3 programming
Data Generation Using augmentation technique necessary for increasing language.
Transformers the robustness of models by supplying training > Basic understanding of machine learning and
data. In this course, you’ll explore the use of deep learning concepts and pipelines.
transformers for synthetic tabular data generation. > Experience building machine learning models
> Learn More with tabular data.
> Basic understanding of language modeling and
transformers.
Techniques for Learn techniques that can take your RAG system > Familiarity working with LLM-based
Improving the from an interesting proof of concept to a serious applications
Effectiveness of RAG asset. > Familiarity with RAG pipelines
Systems
> Learn More
Assemble a Simple In this course, you’ll step through the “Assemble A Windows or Linux computer with the ability
Robot in NVIDIA Isaac a Simple Robot” tutorial to rig a two-wheel mobile to install Omniverse Launcher and Omniverse
Sim™ robot in a live NVIDIA Isaac Sim GPU environment. applications; internet bandwidth sufficient
to support the Isaac Sim client/server stream
> Learn More
(performance will vary).
Build Beautiful, Custom Experience the NVIDIA Omniverse development Basic familiarity with Python (helpful, not
UI for 3D Tools on platform for builders and creators of virtual worlds. required). Suggested materials to satisfy
NVIDIA Omniverse Become a master in UI with a deep dive into NVIDIA prerequisites: The Python Tutorial.
Omniverse Kit’s powerful omni.ui suite of tools and
frameworks. In this self-paced course, you’ll build
your own custom UI for workflows in Omniverse
with hands-on exercises.
> Learn More
Building a 3D Product In this hands-on lab, you’ll unlock the power of Intermediate Python
Configurator With OpenUSD to build a real-time configurator in experience, including object-
OpenUSD and NVIDIA Omniverse. Along the way, you’ll learn oriented programming and
Omniverse about workflows, asset considerations, and USD libraries.
composition concepts that you can apply directly to
your own development process.
> Learn More
Back
21 | OCT24 | NVIDIA Training Course Catalog
Online, Self-Paced Courses for Developers | Graphics and Simulation
Creating and Extensions are one of the building blocks of Extensions Basic programming skills and familiarity with
Customizing an are one of the building blocks of Omniverse Kit- Python, terminal commands, Github, and
Omniverse Extension based applications, allowing you to customize your computer graphics are useful but not required.
application with functionality and interactivity.
In this hands-on lab, you’ll create an extension,
customize it to make an interactive user interface,
and learn how to extract Omniverse application
commands to code the extension’s functionality.
> Learn More
Developing an To build applications from scratch, NVIDIA offers Basic programming skills, and familiarity with
Omniverse Kit-Based Omniverse Kit SDK and free templates to build Python, terminal commands, Github, and
Application starter applications that can be easily customized and computer graphics are useful, but not required.
extended. This course provides the fundamentals of
using Kit SDK to develop such applications.
> Learn More
Easily Develop Get hands-on experience with NVIDIA Omniverse— A basic understanding of computer graphics
Advanced 3D Layout the platform for connecting and creating physically concepts–such as vertices, meshes, and RGB
Tools on NVIDIA accurate, 3D virtual worlds. See how easy it is values–and an understanding of fundamental
Omniverse to create your own custom scene layout tools in programming concepts in Python like functions,
Omniverse Code with a few lines of Python script. In loops, dictionaries, and arrays.
this self-paced course, you’ll build your own custom
scene layout in Omniverse with hands-on exercises
in Omniverse Code and Python.
> Learn More
Fundamentals In this lab, we’ll cover the fundamentals of working An understanding of fundamental programming
of Working With with Universal Scene Description (OpenUSD). concepts in Python 3 such as functions, loops,
OpenUSD You’ll learn how to use USD for nondestructive dictionaries, and arrays.
workflows, how layers can help with ease and speed
of scene composition, and how to use USD for data
separation and reuse it to accelerate 3D workflows
in industrial use cases.
> Learn More
Back
22 | OCT24 | NVIDIA Training Course Catalog
Online, Self-Paced Courses for Developers | Graphics and Simulation
How to Build a Learn how to take advantage of Universal Scene Intermediate Python experience, including
Native OpenUSD XR Description (OpenUSD) to accelerate your extended object-oriented programming and libraries.
Application reality (XR) development and enhance visual fidelity
like never before. This session will equip you with the
skills and tools necessary to build, customize, and
stream your own OpenUSD native XR applications
using NVIDIA Omniverse and NVIDIA CloudXR..
> Learn More
How to Build OpenUSD This lab introduces the basics of the NVIDIA Intermediate Python experience, including
Applications for Omniverse development platform. You’ll learn how object-oriented programming and libraries.
Industrial Digital Twins to get started building 3D applications and tools
that deliver the functionality needed to support
industrial use cases and workflows for aggregating
and reviewing large facilities such as factories,
warehouses, and more.
> Learn More
Introduction to Robotic Robotic automation has enjoyed great success in > Intermediate knowledge and general comfort
Simulations in NVIDIA recent years with increasing hardware capabilities with Python 3. This includes familiarity with
Isaac Sim driving innovation in simulation and machine learning. functions, classes, and basic design patterns.
In this course, we introduce you to Isaac Sim, NVIDIA > Comfort with NumPy arrays and basic matrix
Omniverse’s solution for simulation and robotics. operations.
You’ll learn how to tap into the simulation loop of a 3D > A Windows or Linux machine with NVIDIA
engine and initialize experiments with objects, robots, Omniverse and the Omniverse Streaming
and physics logic. This can be done programmatically Client app.
using Omniverse Kit and Pixar USD commands, but
the course will use Isaac Sim Core to wrap these
low-level operations in an object-oriented fashion. By
the end of the course, you’ll be able to simulate and
control NVIDIA JetBot™ and Franka Emika robots and
coordinate them together to perform a handoff.
The skills covered in this course are direct
prerequisites for working with Isaac Gym and create a
good starting point for exploring Isaac Sim and other
Omniverse applications. The course is great for those
interested in 3D scene specification and robotic
simulation, but it’s also useful for researchers looking
to expand their toolkits and seasoned developers
interested in exploring design patterns for Omniverse
Kit development.
> Learn More
Back
23 | OCT24 | NVIDIA Training Course Catalog
Online, Self-Paced Courses for Developers | Graphics and Simulation
Learn OpenUSD: This is the seventh course in the Learn OpenUSD: An understanding of fundamental programming
An Introduction to Foundations curriculum, where we’re introducing concepts in Python 3, such as functions, loops,
Strength Ordering the concept of strength ordering. This course dictionaries, arrays, and Python 3-related
serves as a primer for strength ordering, often libraries.
referred to by the acronym that governs its rules
(LIVRPS). Strength ordering is an important concept
for OpenUSD that beginners to be aware of and
will be covered again in-depth in more advanced
modules of this course.
> Learn More
Learning About Stages, This is the first course in the Learn OpenUSD: An understanding of fundamental programming
Prims, and Attributes Foundations curriculum, where we’re introducing concepts in Python 3, such as functions, loops,
essential concepts, vocabulary, and Python best dictionaries, arrays, and Python 3-related
practices for OpenUSD. libraries.
In this course, we’ll introduce fundamental terms
in OpenUSD and get hands-on practice with their
implementation in Python.
This course is designed for beginners and those
with some experience in 3D graphics and OpenUSD.
> Learn More
Learn OpenUSD: This is the sixth course in the Learn OpenUSD: An understanding of fundamental programming
Setting Up Basic Foundations curriculum, and it covers basic concepts in Python 3, such as functions, loops,
Animations animation concepts in OpenUSD. dictionaries, arrays, and Python 3-related
libraries.
In this course, we’ll examine how to animate prim
properties using OpenUSD concepts like timeCode
and timeSample.
> Learn More
Learn OpenUSD: This is the fourth course in the Learn OpenUSD: An understanding of fundamental programming
Traversing Stages Foundations curriculum, where we introduce concepts in Python 3, such as functions, loops,
essential concepts, vocabulary, and Python best dictionaries, arrays, and Python 3-related libraries.
practices for OpenUSD.
In this course, we’ll introduce several methods for
efficiently traversing an OpenUSD stage and get
hands-on practice with their implementation in Python.
> Learn More
Back
24 | OCT24 | NVIDIA Training Course Catalog
Online, Self-Paced Courses for Developers | Infrastructure
Learn OpenUSD: This is the fifth course in the Learn OpenUSD: An understanding of fundamental programming
Understanding Model Foundations curriculum, and we’re discussing concepts in Python 3, such as functions, loops,
Kinds model kinds. dictionaries, arrays, and Python 3-related libraries.
In this course, we’ll explore the types of model kinds
and how they can be used to create an efficient
model hierarchy in OpenUSD.
> Learn More
Learn OpenUSD: Using This is the third course in the Learn OpenUSD: An understanding of fundamental programming
Attributes Foundations curriculum, where we introduce concepts in Python 3, such as functions, loops,
essential concepts, vocabulary, and Python best dictionaries, arrays, and Python 3-related libraries.
practices for OpenUSD. In this course, we will expand
on the knowledge introduced in the Learning About
Stages, Prims, and Properties course to explore
attributes more, including how to add, retrieve, and
leverage attributes in OpenUSD workflows.
> Learn More
Learn OpenUSD: This is the second course in the Learn OpenUSD: An understanding of fundamental programming
Working With Prims Foundations curriculum, where we introduce concepts in Python 3, such as functions, loops,
and Default Schemas essential concepts, vocabulary, and Python best dictionaries, arrays, and Python 3-related
practices for OpenUSD. libraries.
In this course, we will discuss what it means to
create a prim without a schema and review pre-built
default schemas that already exist in OpenUSD.
> Learn More
Synthetic Data How much data is enough? This is a common > Intermediate understanding of Python
Generation for Training question when fine-tuning or training computer (including classes, objects, and decorators):
Computer Vision vision models. In cases where data collection is a learn about this topic from the Python.org
Models limiting factor, we can use synthetic data! NVIDIA tutorials
Omniverse Replicator streamlines synthetic > Basic understanding of Machine Learning and
data generation (SDG) using 3D assets into a Deep Learning concepts and pipelines:learn
single application, with the ability to modify the about this topic from the “Deep Learning
appearance and format of the data. This lab Demystified” video
highlights one of the ways deep learning tools and
Omniverse can be used together to streamline deep
learning workloads.
> Learn More
Back
25 | OCT24 | NVIDIA Training Course Catalog
Online, Self-Paced Courses for Developers | Infrastructure
Infrastructure
Back
26 | OCT24 | NVIDIA Training Course Catalog
Instructor-Led Workshops for Administrators | AI and Data Science
Cluster Administration
Ethernet Cumulus
Cumulus Linux: In this hands-on private training, you’ll learn about None.
Private Workshop NVIDIA Cumulus OS architecture, installation,
configuration, operation, and management of
Cumulus Linux running on NVIDIA switches.
> Learn More
NVIDIA Cumulus Linux: This course focuses on how to build and operate a None.
Customized Advanced state-of-the-art data center or storage fabric with
Training emphasis on troubleshooting. The course covers
advanced topics such as filtering, quality of service
(QoS), Ethernet VPN multihoming (EVPN-MH),
monitoring, and active testing.
> Learn More
Back
27 | OCT24 | NVIDIA Training Course Catalog
Instructor-Led Workshops for Administrators | InfiniBand
InfiniBand
InfiniBand Customized In this course, you’ll learn about InfiniBand Network administrators and IT professionals
Course architecture and how to manage, monitor, and that need to install, configure, manage, monitor,
troubleshoot your InfiniBand network. and troubleshoot the configuration and
performance of InfiniBand networks.
> Learn More
InfiniBand Professional In this course, you’ll learn about InfiniBand and None.
Customized Training Cumulus architecture and how to manage, monitor,
and troubleshoot triad deployment-based networks.
> Learn More
NVIDIA DGX
NVIDIA DGX This course provides an overview of NVIDIA DGX System and network administrators and IT
H200/H100/A100 systems, tools for in-band and out-of-band professionals that need to configure and verify
Administration: Private management, NGC, the basics of running workloads, the configuration and performance of DGX
Workshop and specific management tools and command-line systems.
interface (CLI) commands. In addition, this course
includes content on Multi-Instance GPU (MIG),
managing storage, performance validation, and other
system management tools and concepts.
> Learn More
NVIDIA DGX This course provides an overview of DGX System and network administrators and IT
H200/H100/A100 systems and tools for in-band and out-of-band professionals that need to configure and verify
Administration: Public management, the basics of running workloads, the configuration and performance of DGX
Workshop specific management tools, and CLI commands. systems.
> Learn More
Back
28 | OCT24 | NVIDIA Training Course Catalog
Instructor-Led Workshops for Administrators | NVIDIA DGX
NVIDIA DGX BasePOD™ This course provides an overview of DGX BasePOD System and network administrators and IT
Administration: Private components and related processes, including professionals that need to configure and verify
Workshop the NVIDIA DGX A100 system, InfiniBand and the configuration and performance of DGX A100
Ethernet networks, tools for in-band and out-of- clusters.
band management, NGC, the basics of running
workloads, and specific management tools and CLI
commands. It includes instructions for managing
vendor-specific storage per the architecture of your
specific DGX BasePOD solution.
> Learn More
NVIDIA DGX This course is designed to help IT professionals System and network administrators and IT
SuperPOD™ successfully administer all aspects of a DGX professionals that need to configure and verify
Administration: SuperPOD cluster, including compute, storage, and the configuration and performance of DGX
Private Workshop networking. SuperPOD clusters.
> Learn More
Virtualization
NVIDIA AI Enterprise This course covers the platform and solution System administrators and IT professionals that
Administration: overview, hardware and software architecture, need to install, configure, manage, monitor, and
Public Bootcamp deployment options, licensing, temporal and spatial troubleshoot the configuration and performance
GPU partitioning, scaling, comprehensive validation, of their NVIDIA AI Enterprise solution.
management, maintenance, monitoring, and
troubleshooting.
> Learn More
Back
29 | OCT24 | NVIDIA Training Course Catalog
Self-Paced Courses for Administrators | AI and Data Science
AI for All - From Basics This introductory course provides invaluable insights None
to Gen AI Practice into the evolving landscape of AI. Whether you’re a
seasoned professional or just beginning your journey
into AI, this course is essential for staying ahead in
today’s rapidly evolving technological landscape
> Learn More
NVIDIA AI Enterprise This course covers the platform and solution To gain the most value from this course,
Administration overview, hardware and software architecture, the target audience should have a working
deployment options, licensing, temporal and spatial knowledge in the following domains:
GPU partitioning, scaling, comprehensive validation, > Data Center Infrastructure: Servers, Storage,
management, maintenance, monitoring, and Networking, GPUs, Operating Systems.
troubleshooting. > Virtualization: VMware vSphere.
> Learn More > Containerization: Docker.
Cluster Administration
Back
30 | OCT24 | NVIDIA Training Course Catalog
Self-Paced Courses for Administrators | AI and Data Science
Ethernet
Network This course provides the required set of skills to > Basic understanding of Ethernet network
Administration With configure and manage NVIDIA Ethernet switch principles.
the NVIDIA Onyx™ systems. You’ll learn in depth layer 2 configurations > Basic understanding of switching and
Switch System such as VLAN, STP, LAG, and MLAG, as well as how to routing concepts.
configure layer 3 features such as BGP.
> Learn More
RDMA Over Converged In this course, you’ll learn what RoCE is, how it works, Basic understanding of networking concepts and
Ethernet (RoCE) From the different network types RoCE can run over, and the Open Systems Interconnection (OSI) model.
A to Z how to configure RoCE for each network type.
> Learn More
InfiniBand
InfiniBand Essentials This self-paced course covers the fundamental first General understanding of networking concepts
steps into the world of InfiniBand. If you’re looking and principles.
to become more familiar with InfiniBand’s benefits,
uses, architecture layers, and management
concepts, this is the best place to start.
> Learn More
Back
31 | OCT24 | NVIDIA Training Course Catalog
Self-Paced Courses for Administrators | Management
InfiniBand Professional This course covers the fundamentals of the General understanding of networking concepts
InfiniBand technology from a usability point of and principles.
view and builds on the details of the InfiniBand
architecture specification. You’ll learn how to install,
configure, manage, troubleshoot, and monitor your
InfiniBand network.
> Learn More
Management
Data Center Learn about NVIDIA Unified Fabric Manager (UFM) Understanding of InfiniBand fabrics and
Management Made and its capabilities, advantages, and components management concepts
Easy With through a set of interactive learning units, videos,
NVIDIA UFM® and simulators.
> Learn More
NVIDIA License System NVIDIA License System (NLS) is a new licensing > Basic understanding of virtual appliances
solution to support the continued expansion of the installation and setup.
NVIDIA enterprise software portfolio. This course > Familiarity with web/cloud-based applications.
will help you to learn about NLS and how you can > Familiarity with NVIDIA products like virtual
move from your existing licensing solution to NLS. GPU (vGPU) and NVIDIA AI Enterprise.
> Learn More
Network
Ansible Essentials for In this course, you’ll explore a variety of Ansible > Basic Linux administration.
Network Engineers modules and write playbooks specifically adapted > General understanding of networking concepts
to modern data centers. This course includes an and principles.
exclusive hands-on lab environment and exercises
to practice real-world scenarios in real cloud
environments.
> Learn More
Back
32 | OCT24 | NVIDIA Training Course Catalog
Self-Paced Courses for Administrators | Network
MLXlink and MLXcables In this course, you’ll learn about the MLXlink and Good technical background and understanding
Debug Tools MLXcables debug tools. These debug tools are of networking hardware.
used for both basic link troubleshooting and for
analyzing the more complex link characteristics.
> Learn More
NVIDIA BlueField® DPU Learn the basic concepts of BlueField DPUs as a > Basic knowledge and experience in networking
Administration platform for accelerated data center computing. concepts and principle.
> Learn More > Basic knowledge and experience in Linux
administration.
RDMA
The Fundamentals of This course allows C programmers to dive into Understanding of C/C++ programming.
RDMA Programming the RDMA programming world without requiring
previous experience in networking or RDMA
programming. We’ve also added tips and tricks, as
well as do’s and don’ts, so the skills you acquire will
truly serve you when you need them.
> Learn More
Back
33 | OCT24 | NVIDIA Training Course Catalog
Certifications
Certifications
Certification Name Description Prerequisites
NVIDIA-Certified This is an entry-level credential that validates the A basic understanding of data center
Associate: AI foundational concepts of AI computing related to infrastructure.
Infrastructure and infrastructure and operations. The exam is online
Operations and proctored remotely, includes 50 questions, and
has a 60-minute time limit.
> Learn More
NVIDIA-Certified An entry-level credential that validates the A basic understanding of generative AI and large
Associate: Generative foundational concepts for developing, integrating, and language models
AI Large Language maintaining AI-driven applications using generative
Models AI and large language models (LLMs) with NVIDIA
solutions. The exam is online and proctored remotely,
includes 50 questions, and has a 60-minute time limit.
> Learn More
NVIDIA-Certified This is an intermediate level certification that A thorough understanding of data center
Professional: InfiniBand validates core concepts for designing, deploying, infrastructure and networking.
and managing NVIDIA InfiniBand fabrics. The exam
is online and remote proctored with 40 questions
and a time limit of 90 minutes for completion.
> Learn More