Question bank
Oracle Cloud Infrastructure 2026 Data Science Professional
41 Oracle Cloud Infrastructure 2026 Data Science Professional questions with the correct answer and a written explanation, organised across 5 exam domains.
- 41 questions
- 5 domains
- Foundation level
certilumQuestion index
Jump straight to any question. Answers and explanations are inline.
Apply MLOps Practices
8 questions in this domain
- Question 1Single Choice
Which OCI Data Science resource is used to logically group the artifacts of an ML project, such as notebook sessions, jobs, pipelines, and model deployments?
- AModel catalog
- BProjectCorrect answer
- CJob
- DModel deployment
Why this is the answer
A project is the logical container in OCI Data Science for organizing related ML resources such as notebook sessions, jobs, pipelines, and model deployments.
- Question 2Multiple Select
Which statements correctly describe characteristics of the OCI Data Science model catalog? Select all that apply.
- AIt stores model artifacts and associated metadataCorrect answer
- BIt supports registering and versioning models for reuse and managementCorrect answer
- CIt directly executes training jobs on dedicated infrastructure
- DIt provides registered models to model deployments for online servingCorrect answer
Why this is the answer
The model catalog stores, registers, and versions model artifacts. Model deployments use models from the catalog. Training is performed by jobs or pipelines, not by the model catalog.
- Question 3Interactive
A data scientist runs the command presented in the code block. Which OCI Data Science resource is created by this command?
- ANotebook session
- BJobCorrect answer
- CModel deployment
- DPipeline
Why this is the answer
The oci data-science job create command creates a job definition for running ML workloads on OCI Data Science.
- Question 4Single Choice
Which OCI mechanism is recommended for a Data Science job or notebook session to access OCI services without requiring long-lived user API keys in the code?
- AResource principalCorrect answer
- BCustomer-managed key
- CStatic API signing key
- DObject Storage namespace
Why this is the answer
Resource principals allow OCI Data Science resources to call OCI services based on policies, avoiding the need to embed long-lived user credentials in code.
- Question 5Multiple Select
Which practices are core to applying MLOps on OCI Data Science? Select all that apply.
- AAutomating experiment training through Data Science Jobs or PipelinesCorrect answer
- BStoring versioned model artifacts in the model catalogCorrect answer
- CManually copying model files to instances after every retraining run
- DMonitoring deployed models to detect drift and performance degradationCorrect answer
Why this is the answer
MLOps emphasizes automation, versioning, reproducibility, and monitoring. Manually copying model files is not a scalable or auditable MLOps practice.
- Question 6Interactive
The code block uses the OCI Python SDK from a notebook session. What does this code create in OCI Data Science?
- AModel catalog entryCorrect answer
- BPipeline run
- CJob run
- DModel deployment
Why this is the answer
The Python SDK passes CreateModelDetails to create_model, which creates a model entry in the OCI Data Science model catalog.
- Question 7Single Choice
What is the main purpose of a Data Science pipeline in OCI Data Science?
- AOrchestrating and automating multi-step machine learning workflowsCorrect answer
- BProviding a browser-based JupyterLab environment
- CCreating an Object Storage bucket for training data
- DPublishing custom metrics to the Monitoring service
Why this is the answer
A Data Science pipeline chains jobs and steps into an orchestrated, repeatable workflow, unlike a notebook session or storage service.
- Question 8Multiple Select
Which configuration items are typically needed when creating an OCI Data Science notebook session? Select all that apply.
- ACompartmentCorrect answer
- BProjectCorrect answer
- CCompute shapeCorrect answer
- DModel deployment endpoint URL
Why this is the answer
Notebook sessions are created inside a compartment and project and require a compute shape. Model deployment endpoints are not required to create a notebook session.
Design and Set up Data Science Workspace
7 questions in this domain
- Question 9Single Choice
What is the primary resource in OCI Data Science used to logically organize notebook sessions, jobs, and models?
- AProjectCorrect answer
- BNotebook Session
- CModel Deployment
- DData Flow Application
Why this is the answer
A project is the main logical container in OCI Data Science. It organizes related models, notebook sessions, and jobs so they can be managed and collaborated on together.
- Question 10Multiple Select
What must be configured or created before a data scientist can launch a notebook session in OCI Data Science?
- ACreate an OCI projectCorrect answer
- BConfigure a virtual cloud network and subnet for the notebook sessionCorrect answer
- CGrant IAM policies so the Data Science service can use network resourcesCorrect answer
- DSet up an on-premises database for all training data
Why this is the answer
Before launching a notebook session, you must have a project, a virtual cloud network or subnet for the session, and IAM policies that allow the Data Science service to use resources in OCI.
- Question 11Interactive
Review the OCI CLI command in the code block. Which OCI Data Science resource is created by executing this command?
- AA projectCorrect answer
- BA notebook session
- CA model catalog entry
- DA data pipeline
Why this is the answer
The command uses the OCI Data Science CLI service to create a project with the display name Customer_Churn_Project in the specified compartment.
- Question 12Single Choice
Which statement accurately describes a notebook session in OCI Data Science after the workspace and project have been set up?
- AA managed JupyterLab environment running on a compute resourceCorrect answer
- BA static dashboard used only for viewing model metrics
- CA desktop application installed on the data scientist's laptop
- DA SQL query engine used exclusively for data preparation
Why this is the answer
A notebook session in OCI Data Science is a fully managed JupyterLab environment that runs on a compute resource and provides an interactive space for developing machine learning models.
- Question 13Multiple Select
Which of the following are core OCI Data Science resource types that a data scientist can work with in a project?
- AProjectsCorrect answer
- BNotebook sessionsCorrect answer
- CThe model catalogCorrect answer
- DBig Data Service clusters
Why this is the answer
Projects, notebook sessions, and the model catalog are all part of OCI Data Science. Big Data Service is a separate OCI service and is not a Data Science workspace resource.
- Question 14Interactive
Read the IAM policy in the code block. What does this policy allow the OCI Data Science service to do?
- AUse virtual network resources in the tenancyCorrect answer
- BDelete any VCN in the tenancy
- CRead all data in Object Storage
- DCreate Data Science projects in any region
Why this is the answer
The policy allows the OCI Data Science service to use VCNs, subnets, and related network resources in the tenancy so that notebook sessions can run securely.
- Question 15Single Choice
Which OCI construct should be used to scope Data Science projects and notebook sessions for access control and governance?
- ACompartmentCorrect answer
- BRegion
- CVCN
- DAvailability Domain
Why this is the answer
Compartments are logical containers used in OCI to organize and control access to resources. Data Science artifacts should be created in a compartment appropriate for the team and workload.
Implement end-to-end Machine Learning Lifecycle
18 questions in this domain
- Question 16Single Choice
Which OCI Data Science artifact acts as the container for resources created for one machine-learning business use case?
- AProjectCorrect answer
- BCompartment
- CObject Storage bucket
- DNotebook session
Why this is the answer
A Project is the logical container in OCI Data Science that groups notebook sessions, jobs, models and deployments. Compartments are used to organize OCI resources at a broader tenancy level.
- Question 17Multiple Select
Which of the following are core components used when implementing an end to end ML workload on OCI Data Science?
- ANotebook sessionCorrect answer
- BModel catalogCorrect answer
- CModel deploymentCorrect answer
- DData Flow application
Why this is the answer
Notebook sessions support interactive development, the model catalog stores model artifacts, and model deployments serve predictions. Data Flow applications are not an OCI Data Science component.
- Question 18Interactive
Look at the Python code block. What type of resource is created by the client call?
- AProjectCorrect answer
- BModel catalog
- CNotebook session
- DData Science job
Why this is the answer
The request object in the code describes a project and invokes client.create_project. Therefore the returned response contains a Project resource in OCI Data Science.
- Question 19Single Choice
Which statement best describes a notebook session in OCI Data Science?
- AManaged JupyterLab environment with attached computeCorrect answer
- BSQL editor for Autonomous Database
- CData integration pipeline designer
- DContainer orchestration platform
Why this is the answer
An OCI Data Science notebook session is a managed JupyterLab environment with a selected compute shape and block storage. It is used for interactive exploration and model development.
- Question 20Multiple Select
Which items do you explicitly configure when you create a notebook session in OCI Data Science?
- ACompute shapeCorrect answer
- BBlock storage sizeCorrect answer
- CSubnetCorrect answer
- DModel algorithm
Why this is the answer
When creating a notebook session, the user selects a compute shape, block storage size, and subnet. Model accuracy is a training evaluation metric, not a notebook session configuration value.
- Question 21Interactive
The code block registers an ML artifact. What kind of OCI Data Science resource is created?
- AModelCorrect answer
- BProject
- CNotebook session
- DData Flow application
Why this is the answer
CreateModelDetails and client.create_model add a model to the OCI Data Science model catalog. The resource created by the API call is therefore a Model.
- Question 22Single Choice
Which OCI Data Science feature persists trained model artifacts and metadata for later reuse?
- AModel catalogCorrect answer
- BProject
- CJob
- DContainer registry
Why this is the answer
The model catalog stores model artifacts and associated metadata so models can be reused, evaluated, and deployed later. Projects organize resources, while jobs execute code.
- Question 23Multiple Select
Which model-related tasks are supported by OCI Data Science in the end-to-end ML lifecycle?
- ATrain in notebook sessionCorrect answer
- BRun as a repeatable jobCorrect answer
- CDeploy as a REST endpointCorrect answer
- DPurchase database licenses
Why this is the answer
In OCI Data Science a model can be trained in a notebook, run in a job, and served through a deployment. Buying database licenses is unrelated to the model lifecycle.
- Question 24Interactive
Look at the code block. What type of resources are returned by the API request?
- AModel deploymentCorrect answer
- BProject
- CJob
- DConda environment
Why this is the answer
The code calls list_model_deployments, so the returned collection contains model deployments. This method does not list projects, jobs, or conda environments.
- Question 25Single Choice
Which statement best describes a data science job in OCI Data Science?
- AManaged and repeatable execution of codeCorrect answer
- BReal-time model prediction endpoint
- CPersistent storage for training data
- DVisual ETL flow canvas
Why this is the answer
Data Science jobs provide managed, repeatable execution of ML code without an interactive notebook. They are commonly used for scheduled and production workloads.
- Question 26Multiple Select
Which of the following can be created inside an OCI Data Science project?
- ANotebook sessionsCorrect answer
- BData Science jobsCorrect answer
- CModel deploymentsCorrect answer
- DOracle Analytics Cloud workbook
Why this is the answer
Projects group all OCI Data Science resources created for one use case, including notebook sessions, jobs, pipelines, models, and deployments. Oracle Analytics Cloud is a separate analytics service.
- Question 27Interactive
The code block configures a client without embedding user credentials. Which principal is being used?
- AResource principalCorrect answer
- BUser API key
- CFederation user
- DDatabase user
Why this is the answer
The code uses get_resource_principals_signer, which returns the OCI resource principal signer. Resource principals allow a Data Science job to make API calls without user API keys.
- Question 28Single Choice
Which OCI service should a deployed model use to keep database credentials secure?
- AOCI VaultCorrect answer
- BIn the model artifact
- CIn a public Git repository
- DIn plain text logs
Why this is the answer
Secrets should be managed in OCI Vault and referenced by OCID. Storing credentials in artifacts, repositories, or logs creates a security risk.
- Question 29Multiple Select
What must be provided when deploying a registered model as an online endpoint in OCI Data Science?
- AA registered model in the catalogCorrect answer
- BCompute shape for inferenceCorrect answer
- CLogging and network settingsCorrect answer
- DSource code repository URL
Why this is the answer
A model deployment requires a model from the catalog, compute shape, network, and logging settings. A source-code repository is not part of the deployment configuration.
- Question 30Interactive
What effect does the API operation shown in the code block have on the model deployment?
- AActivates the deploymentCorrect answer
- BDeletes the deployment
- CCreates the deployment
- DStops the deployment
Why this is the answer
The activate_model_deployment operation resumes a stopped model deployment and makes it available to serve prediction requests. It does not create or delete the deployment.
- Question 31Single Choice
Which statement correctly describes a model deployment in OCI Data Science?
- AManaged endpoint for online predictionsCorrect answer
- BAutomated training job scheduler
- CData transformation service
- DNotebook kernel manager
Why this is the answer
A model deployment hosts a model behind a managed HTTP service. It provides low-latency online predictions to client applications.
- Question 32Multiple Select
Which activities normally happen before a model is deployed in OCI Data Science?
- AExploratory data analysisCorrect answer
- BModel training and tuningCorrect answer
- CModel evaluation and selectionCorrect answer
- DIncrease deployment instances
Why this is the answer
Before deployment, data scientists explore data, train multiple candidates, tune hyperparameters, and evaluate results. Increasing instances is a deployment scaling activity that occurs after an endpoint exists.
- Question 33Interactive
The code block retrieves metadata about one resource. Which resource is requested?
- AModelCorrect answer
- BProject
- CJob
- DNotebook session
Why this is the answer
The client.get_model call retrieves a single registered model from the model catalog. The response data is a Model object with its metadata.
OCI Data Science - Introduction and Configuration
4 questions in this domain
- Question 34Single Choice
Which statement correctly describes the Oracle Cloud Infrastructure Data Science service for a data science team?
- AA fully managed machine learning platformCorrect answer
- BA relational database administration service
- CA serverless web hosting service
- DA traditional data warehouse
Why this is the answer
OCI Data Science is a fully managed machine learning platform for the complete ML lifecycle, from experimentation to deployment.
- Question 35Multiple Select
Which two statements about how projects support an OCI Data Science workflow are true? (Choose two.)
- AProjects organize notebook sessions and jobsCorrect answer
- BProjects are created to hold only one experiment
- CA project is created before its associated resourcesCorrect answer
- DProjects can contain only model artifacts
Why this is the answer
Projects are core containers in OCI Data Science. They are created first and organize notebook sessions, jobs, and related artifacts throughout the ML workflow.
- Question 36Interactive
Review the shell command in the code block. What is the main result when this command is executed successfully?
- AA project is createdCorrect answer
- BA notebook session is started
- CA model is registered
- DA deployment is provisioned
Why this is the answer
The command creates a Data Science project, which is the parent container needed before adding resources such as notebook sessions and jobs.
- Question 37Single Choice
Which configuration choice determines the amount of compute power allocated to an OCI Data Science notebook session?
- AShapeCorrect answer
- BBucket
- CSubnet
- DDisplay name
Why this is the answer
A shape is selected when creating a notebook session and defines the CPU, GPU, and memory resources assigned to that session.
Use related OCI Services
4 questions in this domain
- Question 38Single Choice
In the context of machine learning operations, what is the main purpose of using the OCI Data Science managed platform in Oracle Cloud?
- ABuilding, training, and deploying machine learning modelsCorrect answer
- BHosting corporate web applications at scale
- CReplacing traditional data warehouses
- DSynchronizing on-premise files with cloud storage
Why this is the answer
OCI Data Science is Oracle's managed service for end-to-end machine learning, including building, training, and deploying models. The other options describe storage, hosting, and file synchronization services.
- Question 39Multiple Select
Which of the following are typical component choices that must be configured when setting up an OCI Data Science solution for a machine learning project?
- AProjectsCorrect answer
- BNotebook sessionsCorrect answer
- CModel deploymentsCorrect answer
- DCloud Shell
Why this is the answer
An OCI Data Science configuration typically includes projects, notebook sessions, and model deployments. Cloud Shell is a browser-based shell environment, not a component of the Data Science service.
- Question 40Interactive
A data scientist runs the displayed code segment in an OCI notebook session during environment configuration. Which OCI service is the code attempting to access?
- AIdentity and Access ManagementCorrect answer
- BObject Storage
- CData Science
- DFunctions
Why this is the answer
The code imports the OCI SDK and creates an IdentityClient using a config file. Calling get_user requests user information from OCI Identity and Access Management.
- Question 41Single Choice
When storing training datasets and model artifacts for an OCI Data Science project, which related OCI service should be used as the primary storage service?
- AObject StorageCorrect answer
- BFile Storage
- CBlock Volume
- DLoad Balancer
Why this is the answer
Object Storage is Oracle Cloud's scalable storage service and is commonly used to hold datasets, notebooks, and model artifacts for OCI Data Science workloads.
Frequently asked questions
Everything about the Oracle Cloud Infrastructure 2026 Data Science Professional question bank and how to use it.