FORWARD-DEPLOYED ENGINEERING / VENDOR COURSES
Know the platform.
Build on it with confidence.
Start with what the pieces are. See how they connect. Build an environment you can explain, then use the official vendor curriculum to prepare for the certification you want.
Try the interactive cloud placement lab- 101Understand
Terminology, familiar comparisons and system diagrams.
- 102Build
Identity, networks, servers, data and model endpoints.
- 103Demonstrate
Failure cases, operating evidence and a study-gap map.
INFRASTRUCTURE + AI / VENDOR COURSE
Understand the GPU. Operate the system.
GPU execution, CUDA, memory, interconnects, Nemotron, NeMo, NIM and the cluster underneath a private AI service.
Start NVIDIAAuthorized data → model request → checked result
NIM / containers · Kubernetes or Slurm · monitoring
GPU + HBM · driver + CUDA · compute/network/storage
Learn these names first
- GPU
- A processor designed for parallel numerical work.
- CUDA
- NVIDIA's platform and programming model for accelerated computing.
- Nemotron / NeMo / NIM
- Model family / development toolchain / supported model-serving microservices.
- NVLink / InfiniBand / Ethernet
- Different communication technologies; choose by the system topology and workload.
Before you startStart with FDE 101. For cluster practice, add Linux, containers and access to an authorized GPU environment.
HANDS-ON WORK
GPU operations and certification workbook
Synthetic diagnosis and read-only hardware inspection. Existing capacity lab runs without a GPU.
Download the lab workbook Open the runnable capacity lab →Your course sequence
- Where does NVIDIA fit in an AI system?25m
- What is a GPU, and why does AI use it?30m
- What is CUDA?50m
- Why does GPU memory matter so much?40m
- How do modern GPU systems work together?35m
- What is Nemotron?35m
- What is NeMo, and when would I use it?40m
- How do NIM and the inference tools fit together?40m
- How do you buy or rent GPU capacity?40m
- 102 · How a GPU becomes available to a workload50m
- 103 · Plan NVIDIA infrastructure certification from evidence40m
CONTINUE TOWARD CERTIFICATION
Choose the role you want to demonstrate.
Sources checked
Blackwell/Blackwell Ultra and Vera Rubin are covered as architectures. Exact product support, regional availability and delivery dates still need current supplier evidence.
Associate · NCA-AIIOAI Infrastructure and Operations
Connections to our course
Accelerated computing and GPU terminology
Software and infrastructure considerations
Continue with the vendor
Complete NVIDIA's official fundamentals training and blueprint review, including broader AI concepts, infrastructure and operations topics. Our self-checks are practice, not the vendor exam.
NVIDIA fundamentals course and exam blueprint ↗Official certification / exam guide ↗Professional · NCP-AIIAI Infrastructure
Connections to our course
System components and deployment planning
Allocation and validation concepts
Continue with the vendor
Real server bring-up, physical-layer work, cluster installation and validation on supported infrastructure. This course's diagrams and calculations do not replace the official workshop or hardware practice.
NVIDIA infrastructure workshop and recommended reading ↗Official certification / exam guide ↗Professional · NCP-AIOAI Operations
Connections to our course
GPU allocation, monitoring and troubleshooting
Role and evidence planning
Continue with the vendor
Hands-on Base Command Manager, Slurm and Kubernetes administration and recovery. The current vendor exam includes a hands-on component; follow its prerequisites and complete real cluster practice.
NVIDIA operations workshop and study guide ↗Official certification / exam guide ↗
Google CloudINFRASTRUCTURE + AI / VENDOR COURSE
From your first project to an AI service.
Projects, regions, zones, Compute Engine, VPC, IAM and the managed model platform—explained through one delivery case.
Start Google CloudGoogle CloudENVIRONMENT MAP- Global VPC
- Regional subnet
- Zonal Compute Engine VM
Learn these names first
What is Google Cloud? Start with the visual 101 →
- Project
- Resource ownership, API usage and access scope.
- Region / zone
- Deployment geography / location within a region.
- Compute Engine
- Google Cloud's virtual-machine service.
- Gemini Enterprise Agent Platform
- The developer platform evolving from Vertex AI; the model is a separate component.
Before you startBasic terminal, HTTP and networking. A dedicated authorized project is needed for the optional live server lab.
HANDS-ON WORK
Build and remove a private Compute Engine server
Original CLI walkthrough. Provisioning uses your approved project and can incur charges; no cloud resources are created by this site.
Download the lab workbook Practise Google Cloud regions and zones →Run the free model response lab →Your course sequence
- 101 · What is Google Cloud, and where does your work live?30m
- 101 · Regions, zones and the network around a server55m
- 102 · Give the application its own identity40m
- 102 · Build and inspect a private Compute Engine server50m
- 102 · Put documents, databases and training data in the right place40m
- 102 · Place Gemini, model endpoints and agents in the system60m
- 103 · Operate, recover and control the cost60m
- 103 · Defend the deployment and plan the certification work40m
CONTINUE TOWARD CERTIFICATION
Choose the role you want to demonstrate.
Sources checked
Current product documentation uses Gemini Enterprise Agent Platform. Exam guides and existing APIs may retain Vertex AI terminology; the course explains both names.
Associate · ACEAssociate Cloud Engineer
Connections to our course
Environment, access and solution planning
Compute, data and operations
Continue with the vendor
Additional official labs for GKE, managed databases, storage administration, organization policy, load balancing and recovery. The course supports selected objectives, not complete exam coverage.
Google Skills · Cloud Engineer learning path ↗Official certification / exam guide ↗Professional · PMLEProfessional Machine Learning Engineer
Connections to our course
AI solution deployment and evaluation planning
Evidence and study-gap review
Continue with the vendor
Model training/tuning, data and ML pipelines, automation, experimentation and the wider ML lifecycle. Complete the official learning path and its labs before assessing readiness.
Google Skills · Machine Learning Engineer path ↗Official certification / exam guide ↗
AWSINFRASTRUCTURE + AI / VENDOR COURSE
Know the account. Build the agent's environment.
Accounts, Regions, Availability Zones, IAM roles, EC2, S3 and Bedrock, with a private-server lifecycle and operating handover.
Start AWSAWSENVIRONMENT MAP- Regional VPC
- Availability Zone subnet
- Private EC2 instance
Learn these names first
What is AWS? Start with the visual 101 →
- Account / role
- Resource boundary / assumable identity with permissions.
- Region / Availability Zone
- Geographic deployment area / isolated location within it.
- EC2 / AMI / EBS
- Virtual server / machine image / block storage.
- Bedrock / AgentCore
- Managed AI capabilities / capabilities for deploying and operating agents.
Before you startBasic terminal, HTTP and networking. An authorized sandbox account and approved CLI identity for live practice.
HANDS-ON WORK
Launch and clean up a private EC2 instance
Original CLI walkthrough with explicit resource IDs. Requires your approved sandbox; creating resources can incur charges.
Download the lab workbook Practise AWS regions and zones →Run the free model response lab →Your course sequence
- 101 · What is AWS, and what is an account?30m
- 101 · Regions, Availability Zones, VPCs and subnets55m
- 102 · Roles, policies and the identity of a running service40m
- 102 · Launch, inspect and remove a private EC2 instance50m
- 102 · S3, databases and evidence for a model40m
- 102 · Bedrock, model access and an agent runtime60m
- 103 · CloudWatch, recovery and the cost of the whole system60m
- 103 · Choose the AWS certification route from your evidence40m
CONTINUE TOWARD CERTIFICATION
Choose the role you want to demonstrate.
Sources checked
The advanced application route uses Generative AI Developer – Professional (AIP-C01). Foundational AI knowledge and ML engineering remain separate study routes.
Associate · SAA-C03Solutions Architect – Associate
Connections to our course
Secure and resilient architecture decisions
Performance, data and cost tradeoffs
Continue with the vendor
Broader service selection, database designs, hybrid networking, resilience and disaster-recovery patterns. Complete official Skill Builder labs and exam preparation.
AWS · Solutions Architect training ↗Official certification / exam guide ↗Foundational · AIF-C01AI Practitioner
Connections to our course
Managed AI, responsible access and evaluation
Continue with the vendor
The complete AI/ML concepts, foundation-model and responsible-AI syllabus. Pair this route with FDE 101 and the vendor's official exam preparation.
AWS · official AI Practitioner preparation ↗Official certification / exam guide ↗Professional · AIP-C01Generative AI Developer – Professional
Connections to our course
Data, model integration and operations foundations
Continue with the vendor
Advanced implementation, orchestration, optimization and production experience across the current blueprint. Our introductory infrastructure course is a prerequisite bridge, not professional exam completion.
AWS · official generative AI exam-prep plan ↗Official certification / exam guide ↗Associate · MLA-C01Machine Learning Engineer – Associate
Connections to our course
Infrastructure, identity and data prerequisites
Continue with the vendor
ML data preparation, model development, SageMaker deployment, workflow automation and monitoring. Follow the official ML route for this additional specialization.
AWS · official ML Engineer preparation ↗Official certification / exam guide ↗
Microsoft AzureINFRASTRUCTURE + AI / VENDOR COURSE
From subscription to AI apps and agents.
Entra tenants, subscriptions, VNets, VMs, managed identities and Microsoft Foundry, connected to current administrator and AI developer study.
Start Microsoft AzureMicrosoft AzureENVIRONMENT MAP- Regional VNet
- Subnet + network security group
- Private Azure VM
Learn these names first
What is Microsoft Azure? Start with the visual 101 →
- Tenant / subscription
- Identity directory / management and billing boundary.
- Resource group
- A lifecycle grouping, not a physical network or region.
- Managed identity
- A workload identity whose credentials the platform manages.
- Microsoft Foundry
- A platform for AI apps and agents; distinct from a model or deployment.
Before you startBasic terminal, HTTP and networking. A dedicated training resource group in an approved subscription for live practice.
HANDS-ON WORK
Create and inspect an isolated Azure VM
Original CLI walkthrough. Uses your approved subscription and may incur charges; cleanup is limited to a dedicated lab resource group.
Download the lab workbook Practise Microsoft Azure regions and zones →Run the free model response lab →Your course sequence
- 101 · Tenant, subscription, resource group: what are they?30m
- 101 · Regions, zones, VNets and private endpoints55m
- 102 · Managed identity, RBAC and data-plane access40m
- 102 · Build an Azure VM without making it public50m
- 102 · Blob Storage, AI Search and the source of truth40m
- 102 · Microsoft Foundry, models and agent tools60m
- 103 · Monitor, recover and hand over an Azure AI service60m
- 103 · Build your AZ-104 and AI-103 study plan40m
CONTINUE TOWARD CERTIFICATION
Choose the role you want to demonstrate.
Sources checked
AI-102 / Azure AI Engineer Associate retired on 30 June 2026. New AI application study points to AI-103, Azure AI Apps and Agents Developer Associate.
Associate · AZ-104Azure Administrator Associate
Connections to our course
Identity, governance and networking
Compute, storage and monitoring foundations
Continue with the vendor
Wider administration, storage configuration, backup/restore, network services and governance labs. Use the complete current Microsoft study guide and course.
Microsoft Learn · Azure Administrator course ↗Official certification / exam guide ↗Associate · AI-103Azure AI Apps and Agents Developer Associate
Connections to our course
AI solution planning and generative/agentic foundations
Continue with the vendor
Computer vision, text analysis, information extraction and the complete implementation labs. This course connects the infrastructure and AI application foundations; it does not claim every AI-103 objective is covered.
Microsoft Learn · Develop AI apps and agents on Azure ↗Official certification / exam guide ↗
AnthropicAGENT-ASSISTED ENGINEERING / CLAUDE CODE
Build with Claude Code.
Learn what a coding agent does, inspect a repository, make a bounded change and test the result. The original course connects the explanation to a complete offline project.
Start Claude CodeGoogle CloudCLOUD FOUNDATIONS / INTERACTIVE PRACTICETwo servers. One shared point of failure?
Place a standby server, then see what a zone outage takes away. Start with the same small application in each cloud.
Location and network are different ideas.
A region is a geographic area. A zone is an isolated deployment area within a region. A subnet is a range of network addresses. Think of choosing a facility separately from choosing a server's network address.
A Google Cloud VPC is a global network. Each subnet belongs to one region, and a VM belongs to one zone in that region.
Zone A and Zone B below are teaching labels, not cloud location codes. Assume a region with two supported zones and enough capacity.
Google CloudPROPOSED PLACEMENTOne selected region
Boxes show resource scope and network attachment. They do not show physical racks or a complete traffic path.
Original local simulation · no cloud account or resource creation. Only placement and the selected outage are modeled. Google Cloud · VPC networks and subnets ↗ Checked 2026-10-20.
Continue with the network lesson, worked example and infrastructure workbook.
Academy access required