NCP-AI Nutanix Certified Professional - Artificial Intelligence NCP-AI 6.10 Free Practice Exam Questions (2025 Updated)

Comprehensive and Detailed Explanation From Exact Extract of Nutanix Artificial Intelligence documents knowledge without any external URL or links:

According to the endpoint creation workflow in the Nutanix Enterprise AI documentation, an endpoint is essentially a deployment of a specific model. Therefore, the prerequisite for creating an endpoint is having a model available in the local registry.

Model Status:The system explicitly requires that the imported Large Language Model (LLM) be in theActive(or Ready) state before it can be selected in the "Create Endpoint" wizard.

Workflow:When an administrator imports a model (whether from Hugging Face or manually), it transitions through statuses like "Processing" (downloading) or "Pending" (waiting for resources). It is only when the status reachesActivethat the model is fully validated, stored, and available to be served as an endpoint. An "Uploaded" status typically refers to a manual file transfer that has not yet been processed or validated for deployment.

Comprehensive and Detailed Explanation From Exact Extract of Nutanix Artificial Intelligence documents knowledge without any external URL or links:

A400 Bad Requesterror typically indicates a client-side issue where the request sent to the server is malformed, has invalid syntax, or is missing required headers. Since the administrator has verified the endpoint is in theActivestate and internal testing via the NAI console returnsSucceeded, the server-side infrastructure and model are functioning correctly.

The most effective resolution is to provide the developers with the exactSample Requestcode generated by the platform.

View Sample Request:In the Endpoint details page, the administrator can click "View Sample Request" (or "View Code").

Content:This feature generates a precise code snippet (often in cURL or Python) that includes the correctly formatted Endpoint URL, the specific API Key, and the required JSON payload structure.

Result:By using this validated code, developers can identify syntax errors in their own code (such as incorrect header formatting or JSON structure) that are causing the 400 error, ensuring immediate connectivity.

Comprehensive and Detailed Explanation From Exact Extract of Nutanix Artificial Intelligence documents knowledge without any external URL or links:

To successfully configure a custom domain with a corporate CA-signed certificate in Nutanix Enterprise AI, the administrator must ensure the following two external prerequisites are met:

DNS Configuration (Option A):AnA recordmust be created in the corporate DNS server that points the fully qualified domain name (FQDN) ai.company.com to the external IP address assigned to the NAI Ingress Gateway (LoadBalancer IP). This ensures that traffic destined for the custom URL reaches the cluster.

Certificate Provisioning (Option C):A validCA-signed TLS certificate(and its corresponding private key) matching the FQDN ai.company.com must be obtained from the corporate Certificate Authority. This certificate is then stored as a Kubernetes Secret (e.g., nai-letsencrypt-certificate or similar custom secret) and referenced by the Ingress Gateway to enable secure HTTPS termination.

Comprehensive and Detailed Explanation From Exact Extract of Nutanix Artificial Intelligence documents knowledge without any external URL or links:

According to the "Large Language Models in Nutanix Enterprise AI" documentation regarding model management, administrators have specific options when importing models that are not pulled directly from Hugging Face or NVIDIA NGC.

Supported Import Sources:For custom LLMs or pre-validated models in "Dark Site" (air-gapped) scenarios, the platform supports importing models fromLocal Storage,S3-compatible Object buckets, orNetwork File System (NFS) file shares.1

Protocol Specifics:Among the file-based protocols listed (SMB, SFTP, FTP, NFS),NFSis the only one explicitly supported for mounting or accessing external file shares to ingest model weights and configuration files into the NAI cluster. This aligns with standard Kubernetes ReadWriteMany (RWX) volume requirements often used in NAI backends.

Comprehensive and Detailed Explanation From Exact Extract of Nutanix Artificial Intelligence documents knowledge without any external URL or links:

To successfully configure and validate the integration, the administrator needs to utilize the connection details provided by the NAI platform.

Validation Workflow:The correct procedure is to retrieve theEndpoint URLand the uniqueAPI Keyfrom the NAI Endpoint details page (often via the "View Sample Request" or "Sample Code" feature).

Configuration:The administrator or developer must then configure the application to send HTTP requests to this specific URL with the API Key included in the header (e.g., as a Bearer token).

Testing:Sending a test request using these exact credentials allows the administrator toconfirm the response meets requirements(schema and content) and validates that the application can successfully authenticate and communicate with the inference service before production rollout.

Comprehensive and Detailed Explanation From Exact Extract of Nutanix Artificial Intelligence documents knowledge without any external URL or links:

When an inference endpoint experiences high latency and timeouts specifically duringpeak usage, it indicates that the current number of model replicas is insufficient to handle the incoming request volume (concurrency).

Scale Out:The standard operational procedure in Nutanix Enterprise AI (which utilizes Kubernetes/KServe under the hood) is toscale outthe endpoint. By analyzing resource metrics (like GPU utilization or request queue depth), the administrator can increase the minimum/maximum replica count for the inference service.

Load Balancing:Adding more replicas allows the load balancer to distribute the inference requests across multiple pods/GPUs, reducing the queue time for each request and resolving the timeout issues.

Why others are incorrect:Restarting the cluster (B) causes downtime and doesn't solve the capacity issue; disabling logging (C) hampers root cause analysis; and limiting keys (A) restricts legitimate business access rather than fixing the performance bottleneck.

Here are the verified answers and detailed explanations for the questions you provided, formatted exactly as requested.

Comprehensive and Detailed Explanation From Exact Extract of Nutanix Artificial Intelligence documents knowledge without any external URL or links:

When deploying the Nutanix Kubernetes Platform (NKP) to host Nutanix Enterprise AI workloads, the underlying operating system of the worker nodes must support both the Kubernetes orchestration layer and the specific NVIDIA GPU drivers required for AI acceleration.

Supported OS:Ubuntu 22.04is a standard, fully supported operating system for NKP GPU nodes. It features extensive compatibility with theNVIDIA GPU Operator, which automates the management of GPU drivers and container toolkits required for NAI.

Other Options:

Rocky Linux 9.5:While NKP supports the Rocky Linux family, version 9.5 is an extremely recent release. Typically, specific minor versions (like 9.2 or 9.3) are validated in documentation. Ubuntu 22.04 is the established "Gold Standard" for broad AI/GPU compatibility in these deployments.

CentOS 8:This OS has reached End of Life (EOL) and is not supported for new modern AI deployments.

Debian 11:While robust, it is less commonly cited as a primary supported OS for the specific combination of NKP and commercial NVIDIA GPU Operators compared to Ubuntu or RHEL/Rocky.

Comprehensive and Detailed Explanation From Exact Extract of Nutanix Artificial Intelligence documents knowledge without any external URL or links:

When a specific endpoint or node is showing 100% GPU utilization while others are underutilized, it indicates that the inference workload is bottlenecked on a single instance of the model. To resolve this latency and uneven distribution, the administrator should rebuild (or update) the endpoint to include multiple instances (replicas).

By increasing the instance count for that specific endpoint, the Nutanix Enterprise AI platform (via the underlying KServe/Knative orchestration) will provision additional pods. These new pods will be scheduled on the other available nodes (which are currently showing moderate usage), thereby distributing the inference requests (load balancing) across multiple GPUs. This effectively lowers the utilization per node and reduces the latency for incoming requests.

Comprehensive and Detailed Explanation From Exact Extract of Nutanix Artificial Intelligence documents knowledge without any external URL or links:

When an endpoint creation fails with a genericFailedstatus and no specific error message is visible in the UI, the Nutanix Enterprise AI documentation explicitly advises administrators to retrieve the detailed status of the Inference Service (isvc) resource using kubectl.

Command:The specific command provided in the troubleshooting procedures is kubectl describe isvc endpoint_name -n nai-admin.

Namespace:User-created endpoints and inference services are typically deployed into thenai-adminnamespace (or the specific tenant namespace), whereas system components reside in nai-system.

Output:The describe command provides the "Conditions" and "Events" sections for the InferenceService, revealing the underlying reason for the failure (e.g., "Insufficient Resources", "ContainerCreationConfigError", or timeout issues), which is essential information for Nutanix Support to diagnose the issue.

Invalid Options:You cannot run kubectl exec against an isvc object (as it is a Custom Resource, not a pod), making options C and D syntactically incorrect.

Comprehensive and Detailed Explanation From Exact Extract of Nutanix Artificial Intelligence documents knowledge without any external URL or links:

Nutanix Enterprise AI enforces strict security protocols regarding API Key visibility.

One-Time View:The documentation warns users to "Store the key securely. You cannot view the key again after you close the API Key Details dialog box".

Recovery Policy:Because the keys are not stored in plain text or retrievable via the dashboard for security reasons, if a key is lost, it cannot be recovered or viewed. The mandatory administrative action is togenerate a new API keyfor the endpoint and update any applications that were using the lost credential.

Comprehensive and Detailed Explanation From Exact Extract of Nutanix Artificial Intelligence documents knowledge without any external URL or links:

Nutanix Enterprise AI provides an inference gateway that is API-compatible with the OpenAI standard, allowing developers to use standard libraries (like the OpenAI Python SDK) to interact with models running on NAI. The supported endpoints that map to this standard include:

/v1/chat/completions: For chat-based text generation (supported).

/v1/embeddings: For generating vector embeddings (supported).

/v1/models: For listing the available models (supported).

The option/v1/systemsisnota supported or valid OpenAI-compatible endpoint. In the OpenAI API schema, "system" refers to a messagerole(e.g., {"role": "system", "content": "..."}), not a standalone URL endpoint. Therefore, this is the unsupported type in the list.

Comprehensive and Detailed Explanation From Exact Extract of Nutanix Artificial Intelligence documents knowledge without any external URL or links:

According to the Nutanix Enterprise AI "User Roles" and privileges documentation, the AI/ML User is the least privileged role that still retains the capability to import Large Language Models (LLMs).

The permissions table explicitly lists "Import an LLM" asYesfor the AI/ML User role. However, a key distinction is that an AI/ML User can only view, manage, and delete the LLMs that they have imported themselves, whereas an AI/ML Admin or Super Admin has broader visibility and management rights over models imported by other users. Since the question asks for theleastprivileged role required to perform the import action,AI/ML Useris the correct answer.

Comprehensive and Detailed Explanation From Exact Extract of Nutanix Artificial Intelligence documents knowledge without any external URL or links:

To install Nutanix Enterprise AI (NAI), specifically when pulling the required container images and Helm charts from the Nutanix repository, authentication is required.2

Access Token Source:The administrator must log in to theNutanix Portal(specifically the Software Downloads or Licenses section) to generate or retrieve theAccess Token(often referred to as the API Key for the registry).3

Usage:This token is then used as the DOCKER_API_KEY or password in the Helm install command (helm install ... --set imagePullSecret.password=...) to authenticate against the ntnx-charts or specific container registry.

Comprehensive and Detailed Explanation From Exact Extract of Nutanix Artificial Intelligence documents knowledge without any external URL or links:

Nutanix Enterprise AI defines specific "Deployment Types" based on the underlying infrastructure platform.

Standalone Deployment:This deployment type is explicitly defined in the documentation as the method for deploying Nutanix Enterprise AI Pro on "any Kubernetes cluster running on bare metal servers, on-prem VMs, andpublic clouds such as Amazon EKS, Azure AKS, or Google GKE".

Differentiation:

GPT-in-a-Box 2.0refers to the full-stack on-premises solution including Nutanix Cloud Infrastructure (NCI) and Hardware.

Cloud Clusteris often a colloquial term or refers to Nutanix Cloud Clusters (NC2), but "Standalone" is the formal deployment SKU/type for bringing your own Kubernetes (BYOK8s) like EKS.

Comprehensive and Detailed Explanation From Exact Extract of Nutanix Artificial Intelligence documents knowledge without any external URL or links:

According to the "Prerequisites" and "Storage Requirements" section of the Nutanix Enterprise AI installation guide, the administrator must provision a specific storage class to handle the shared access required for model files. The documentation explicitly states: "Ensure that you create a ReadWriteMany (RWX) storage class that enables static or dynamic provisioning of NFS shares for persistent volumes."2

This RWX mode is critical because it allows the persistent volume containing the large language models (LLMs) to be mounted as read-write by multiple nodes simultaneously. This ensures that as the inference service scales across different worker nodes, all pods can access the same underlying model data without conflict or duplication. RWO (ReadWriteOnce) would restrict the volume to a single node, which breaks the scalability of the AI cluster.

Comprehensive and Detailed Explanation From Exact Extract of Nutanix Artificial Intelligence documents knowledge without any external URL or links:

According to Nutanix Enterprise AI architecture recommendations for improving Retrieval-Augmented Generation (RAG) performance, when the issue is specifically related to the relevance of the retrieved documents (context) rather than the generation capability of the LLM, the standard solution is to introduce arerankingstep.

Problem Identification:The user explicitly states that when therightdocuments are used, the output is good, but 20% of the time the retrieval is generic or irrelevant. This indicates a precision issue in the retrieval phase, not a generation issue.

Solution (Reranking):Arerank model(or Cross-Encoder) is deployed as a post-retrieval step. It takes the top N documents retrieved by the initial embedding search and re-scores them based on their direct relevance to the specific query, re-ordering the list to ensure the most pertinent information is passed to the LLM. This improves precision without requiring the expensive and time-consuming process of replacing the embedding model or re-indexing the entire database.

Comprehensive and Detailed Explanation From Exact Extract of Nutanix Artificial Intelligence documents knowledge without any external URL or links:

According to the "Sizing Requirements" section of the Nutanix Enterprise AI deployment guide, specific minimum persistent storage values are assigned to each internal component of the NAI control plane.

nai-db Requirements:The nai-db component, which is responsible for storing application data in a PostgreSQL database, explicitly requires a minimum of4 GiBof persistent storage.

Storage Access Mode:This storage is provisioned using a ReadWriteOnce (RWO) access mode, typically backed by Block storage.

Comparison:To differentiate from other components:

The nai-api (Model) component typically requires 20 GiB.

The prometheus-nai component also typically requires 20 GiB.

The 2 TiB requirement mentioned in other contexts usually refers to the shared storage (NFS/Object) required for holding the actual Large Language Model (LLM) weights, not the internal nai-db database.

Comprehensive and Detailed Explanation From Exact Extract of Nutanix Artificial Intelligence documents knowledge without any external URL or links:

According to the Role-Based Access Control (RBAC) definitions within the Nutanix Enterprise AI (NAI) platform, there is a strict segregation of resources to ensure privacy and security in multi-tenant environments.

User Isolation:While anAI/ML Userhas the privileges to create, manage, and delete theirownendpoints and generate API keys for applications they are developing, they do not have global visibility into the private resources of others.

Model Visibility:An AI/ML User can view and deployPre-validated Models(provided globally by the Administrator) orCustom Modelsthat they have imported themselves (if the "Allow Manual Upload" permission is granted).1However, theycannot viewCustom Models that have been imported by other AI/ML Users. These models are scoped to the user who created them to preventing unauthorized access to proprietary or sensitive model weights uploaded by different teams.

Comprehensive and Detailed Explanation From Exact Extract of Nutanix Artificial Intelligence documents knowledge without any external URL or links:

For text-generation models (LLMs), Nutanix Enterprise AI provides a built-in "Test" button in the dashboard that allows users to send sample or custom prompts directly from the UI. However, the documentation explicitly states that this internal Test functionality is not supported for non-text generation endpoints, such as Image Generation or Image Classification models.

To test these specific non-text endpoints, the administrator must access theEndpoint Accesswidget (or "View Sample Request" dialog),copy the Sample API Code(which includes the correct URL and necessary API key), andrun it externallyin a terminal (using curl) or within the client application. This manual execution of the sample code is the verified method for validating non-text endpoints.