USMLE STEP2 Exam Prep
Step2 (Page 27 )

• prompt: the input or instruction given to the model
• completion: the desired response

json 
{"prompt":"Summarize this customer complaint.","completion":"The customer reports a delayed shipment."}

• B: A .txt file containing CSV-style lines is not automatically a valid fine-tuning dataset. The required structure and format must match the selected Bedrock model.
• C: Provisioned Throughput provides dedicated inference capacity; it does not fine-tune a model.
• D: Journals and textbooks might support domain adaptation or pretraining, but they do not provide the prompt-to-answer examples typically required for supervised fine-tuning.

• Updating the CRL: Revokes compromised certificates, but does not fix the CA flaw that allowed them to be created.
• Changing passwords: May be necessary if accounts were compromised, but it does not address the certificate-generation vulnerability.
• Implementing SOAR: Improves security automation and response; it does not remediate the underlying PKI weakness.

• Workspace: Stores notebooks, libraries, jobs, and other project resources.
• Notebooks: Interactive documents containing code, queries, visualizations, and explanations. Common languages include Python, SQL, Scala, and R.
• Clusters: Groups of virtual machines that run Spark workloads. Clusters can be started, stopped, and automatically scaled.
• Data access: Databricks can read from services such as Azure Data Lake Storage, Azure Blob Storage, and Azure SQL.
• Jobs: Automate notebooks or data-processing workflows on a schedule or in response to events.

• Works across SaaS and internal applications: TOTP is widely supported by authentication systems and can be integrated into many applications.
• Device manufacturer agnostic: A hardware token or authenticator application can generate the code without depending on a specific computer or phone manufacturer.
• Offline capability: The code is generated locally using a shared secret and the current time; an internet connection or cellular service is not required.

• Username and password: May work offline but provides weaker authentication and does not meet the intended improvement in security.
• Biometrics: Can depend on device hardware, operating-system support, and enrollment mechanisms.
• SMS verification: Requires cellular connectivity and is vulnerable to threats such as SIM swapping.

• Amazon DynamoDB — Managed NoSQL database, not PostgreSQL.
• Amazon Athena — Serverless service for querying data in S3; intended for analytics, not OLTP.
• Amazon EMR — Managed cluster platform for big-data frameworks such as Hadoop and Spark.

• A — conformant: Incorrect because one explicit requirement is violated.
• B — consistent: “Consistent” is not sufficient where the implementation contradicts a mandatory specification. The RPC mechanism is covered by the architecture, and the implementation differs from it.
• C — consistent: Incorrect reasoning: the RPC mechanism does have a defined architectural requirement, so conformance must be assessed.
• D — non-conformant: Correct because the implementation fails to follow the specified RPC/threading requirement.

• Reviewing and validating generated code
• Checking security, licensing, and quality concerns
• Testing the implementation
• Making the final design and coding decisions

• A: Copilot is not a complete replacement for developers.
• C: Human input, review, and oversight are still required.
• D: Copilot Chat supports many tasks beyond debugging, including explanation, generation, refactoring, and testing assistance.

text 
heartrateheartrateheartrate

• Spark column objects can be multiplied, for example:

python 
  df["heartrate"] * 3

• A DataFrame itself cannot be multiplied, but the expression is referring to a column.
• It is also not primarily a Python syntax error; the code was parsed, then Spark failed while resolving the column name.

sql 
GRANT ALTER ON SCHEMA::schemaA TO User1;

• TRUNCATE TABLE is a DDL operation, not a normal DELETE.
• Having DELETE permission on a table is not sufficient to truncate it.
• The minimum relevant permission is ALTER on the table.
• Granting ALTER on SCHEMA::schemaA allows the user to truncate tables within schemaA.
• It does not grant equivalent permissions on schemaB or schemaC, satisfying least privilege.

• Check whether the data is accurate, complete, and properly prepared.
• Confirm that the pattern is not caused by an outlier, data-quality problem, or modeling issue.
• If the data is valid, treat the result as a potentially valuable new business insight.

• A: A data scientist may help, but expert validation and data-quality checking come first.
• C: Filtering out inconvenient data risks hiding a valid insight.
• D: Einstein’s result should not be accepted without verification simply because it produced the finding.

• S3 Lifecycle rules automate actions such as transitioning objects to different storage classes or deleting old objects.
• S3 bucket policies control access permissions; they do not preserve earlier object versions.
• S3 server-side encryption protects data confidentiality by encrypting objects at rest, but it does not prevent deletion or overwriting.

• Accidental deletion or overwriting ? S3 Versioning
• Unauthorized access ? bucket policies and IAM
• Data encryption ? server-side encryption

• Encryption to preserve confidentiality
• Integrity protection to detect alteration
• Authentication of the server through certificates

• SHA-256: A hashing algorithm; it does not encrypt data for transmission.
• SSL 3.0: Obsolete and insecure. It should not be used.
• AES-256: A strong encryption algorithm, but by itself it does not provide the complete secure communication protocol needed for data in transit.

• Low usage during the first three weeks
• Very high usage during the last week

• High availability: Keeps services running despite failures.
• High latency: Means slow response times; it is not a benefit.
• Load balancing: Distributes traffic across resources but does not specifically address scaling resources up and down.

• R-squared (R²): mede quanto da variação dos valores reais é explicada pelo modelo. É mais usado em regressão, não em classificação.

• Accuracy (acurácia): mede a proporção de previsões corretas em relação ao total. É adequada para a classificação de imagens da questão.

• RMSE (Root Mean Squared Error): mede o tamanho médio dos erros entre valores previstos e reais, penalizando erros maiores. É usado principalmente em regressão.

• Learning rate (taxa de aprendizado): controla o tamanho dos ajustes feitos pelo modelo durante o treinamento. Não é uma métrica de avaliação; é um hiperparâmetro.

• C — Schema.describeSObjects(new String[]{myObject})[0].fields.getMap()

• D — Schema.getGlobalDescribe().get(myObject).getDescribe().fields.getMap()

• A: sObjectType is not used this way, and fieldSet() is not the appropriate method for retrieving all fields.
• B: mySObject.myObject is not valid dynamic Apex syntax.

• Entradas grandes, de até aproximadamente 1 GB.
• Processamentos demorados, que podem levar até cerca de 1 hora.
• O cliente envia a solicitação e consulta o resultado depois, sem precisar manter uma conexão aberta durante todo o processamento.

• Real-time inference: indicada para respostas imediatas, com entradas menores e processamento rápido.
• Serverless inference: adequada para cargas variáveis e solicitações menores; não é ideal para processamento longo e arquivos grandes.
• Batch transform: processa grandes volumes em lote, mas não atende à necessidade de respostas próximas do tempo real.

• Microsoft 365 groups can contain users, but generally can’t contain other groups as members.
• Security groups can contain users and, where supported, other security groups.
• Distribution groups have different nesting restrictions and aren’t equivalent to security groups.
• Dynamic groups can’t be manually populated with members; their membership is calculated from rules.
• A group’s membership type (Assigned or Dynamic) and group type determine whether it can be added to another group.

• A — Custom Button: A button can launch a flow, typically by passing the flow’s URL and record context.
• D — From a Process: A Process can invoke a flow using an Apex action configured for flow invocation. However, this is automation-triggered, not directly accessed by a user.
• B — Quick Action: In current Salesforce functionality, a Flow Quick Action can launch a screen flow directly from a record page. This makes B a valid modern method.
• C — Visual Workflow Launcher: This is not generally listed as a standard access method in current Salesforce terminology; Salesforce commonly uses the Flow component on Lightning pages instead.

• IKE: Negotiates security parameters for IPsec VPNs; it does not identify switch ports.
• VLAN: A logical network segmentation method, not a discovery protocol.
• netstat: Displays local network connections and listening services, but cannot identify the physical switch or port.

• Amazon S3 — Object storage, not a database.
• Amazon DynamoDB — Fully managed NoSQL key-value and document database.
• Amazon Redshift — Managed data warehouse designed for analytics, not primarily transactional MySQL workloads.

• Tactics: The attacker’s overall goal, such as gaining persistence or exfiltrating data
• Techniques: The methods used to achieve that goal
• Procedures: The specific implementation or sequence of actions

• IP addresses and hash values: Useful indicators, but they can change quickly.
• Indicators of compromise: TTPs may help identify IOCs, but their main value is describing behavior.
• Tools: Attackers can replace tools or use legitimate administrative utilities.

• Create and manage catalogs of approved IT services
• Govern and publish infrastructure-as-code (IaC) templates, commonly AWS CloudFormation templates
• Control which users can deploy specific products and configurations
• Apply versioning, permissions, and deployment constraints

• AWS Resource Explorer: Searches and discovers AWS resources; it does not govern IaC templates.
• AWS Organizations: Manages multiple AWS accounts, policies, and consolidated billing.
• AWS Systems Manager: Helps operate and manage deployed resources, such as patching and automation, but is not primarily a catalog for governed IaC products.

• tag=Priv searches the tag field for values beginning with Priv. The wildcard can match the remaining characters, so it matches Privileged.
• tag=Priv would search for the exact tag Priv, not Privileged.
• tag=priv* uses lowercase priv; tag matching is treated as case-sensitive here, so it may not match the tag Privileged.
• tag=privileged uses the wrong capitalization for the exact tag.

• Strongest encryption algorithms: Choose the strongest available encryption and integrity options, typically AES with the largest key size and a secure integrity algorithm.
• Encrypt all traffic: Select tunnel mode rather than transport mode. Tunnel mode encrypts the entire original IP packet, which is appropriate for a site-to-site VPN between branch networks.
• Secret-password authentication: Configure a pre-shared key (PSK) on both VPN concentrators. The key must match at both ends.

• You provision the virtual machine.
• You manage the guest operating system, applications, and configuration.
• Azure manages the physical servers, datacenter, and underlying virtualization infrastructure.

• A: My Domain is required for many SSO configurations, but if SSO works, it is unlikely to be the cause of losing the target record.
• B: HTTP-POST is a valid SAML binding and does not inherently prevent redirection.
• D: An incorrect Federation ID generally causes authentication or user-matching failure—not successful login to the home page.

1. Get Data ? Folder in Power BI Desktop.
2. In Power Query Editor, expand the Attributes column to expose metadata such as hidden status and other file attributes.
3. Remove the Content column (and any other data columns you do not need).

• Name
• Extension
• Date accessed
• Date modified
• Folder Path
• Attributes
• Content

• The VPN gateway is in VirtualNetworkA.
• Gateway transit is configured on the peering associated with the VNet that contains the gateway—here, VirtualNetworkA.
• A peered VNet may use the remote gateway through the appropriate peering option, but this does not automatically update an existing P2S client’s routes.
• The Windows 10 workstation’s P2S VPN client package was created before VirtualNetworkB was peered, so it may not have a route for VirtualNetworkB.

• Uses on-premises data or resources
• Runs components in the cloud
• Can scale across multiple cloud environments when demand increases

• Public cloud: Resources are hosted entirely by a cloud provider.
• Private cloud: Resources are dedicated to one organization, usually in its own environment.
• Hybrid: Combines on-premises and cloud resources.

• A Layer 2 switch checks the destination MAC address in its MAC address table (CAM table).
• If the destination MAC is unknown, the switch cannot select a specific outgoing port.
• It therefore performs unknown-unicast flooding, sending the frame out every other port in that VLAN.
• The frame is not sent back through the port where it arrived.

• Snapshot: Captures a point-in-time state for rollback. It is not an independent running VM and does not protect against failure of the underlying host or storage.
• Convert to dynamic disks: Changes disk allocation behavior; it does not provide redundancy or high availability.
• P2V: Converts a physical machine into a virtual machine. That is a migration process, not an availability solution.

• 25 — Traditional SMTP server-to-server mail transfer; not normally used for secure client submission.
• 110 — POP3, used to retrieve email and typically unencrypted.
• 143 — IMAP, used to access email and typically unencrypted.

• Two VMs
• Distributed across two separate availability zones

• Review the risk register.
• Use the previously defined risk response plan, such as assigning a backup resource, reprioritizing work, or using another qualified engineer.
• Implement the response promptly because the task is on the critical path.

• B: Revising the schedule without first applying the planned response may be premature.
• C: Excluding a required task could compromise the project’s objectives and requirements.
• D: Updating the risk and lessons learned records is appropriate, but it should happen after managing the immediate issue, not as the next primary action.

• Separate accounts/domains create a strong administrative boundary between non-production and production.
• Separate projects provide distinct IAM boundaries, billing, quotas, service accounts, and resources.
• Developers can receive permissions in the development project without automatically gaining access to production.
• Production administrators can be restricted from development resources, supporting separation of duties and least privilege.

• A: Creates excessive project fragmentation without being required by the stated requirement.
• C: Keeps development and production under one account and does not clearly separate administrative duties.
• D: Uses one account and only two projects; it provides some project isolation but weaker organizational separation.

• Create Power Query parameters for the Azure SQL server/database—or at least the database name.
• Use the parameter in the PBIX connection.
• Publish the same PBIX separately to the relevant development, test, and production workspaces.
• Set each dataset’s parameter value in the Power BI service so it points to the corresponding Azure SQL database.

• Development dataset ? DevelopmentDB
• Test dataset ? TestDB
• Production dataset ? ProductionDB

• The same PBIX is used for every lifecycle stage.
• Each stage consumes data from its matching database.
• Deployment pipelines are not required.
• Azure SQL is a cloud source, so gateway administration is generally avoided.
• Only the parameter value differs between published datasets, reducing maintenance.

• Conserve public IPv4 addresses: IPv4 addresses are limited, so many internal devices share a smaller number of public addresses through NAT.
• Reduce direct exposure: Private addresses are not directly routable from the public Internet, which helps reduce unsolicited access to internal devices.

• Assign users or their team an appropriate security role.
• Ensure the role has the required table privileges, such as read, create, write, and append.
• Share the app with that security role in the app designer.

• Dynamics administration center: Used for environment and service administration, not routine app access configuration.
• Manage Roles: Not the specific Dataverse area used to define app and data permissions.
• Security Roles: Define what users can access and which operations they can perform.

1. Identify every Region containing a replica.
2. Schedule deletion of the KMS key in each Region.
3. Use the minimum KMS waiting period: 7 days.

• A — Use the root user: Root access does not eliminate the need to schedule deletion for every regional replica. The problem is the multi-Region structure, not insufficient administrative permissions.

• C — Update the key policy: The policy already grants kms:* to IAM principals. Changing permissions does not delete the key or its replicas.

• D — Disable the key and schedule deletion in 30 days: Disabling can stop cryptographic use, but 30 days is slower than the permitted 7-day waiting period. It also does not address replicas in other Regions.

1. Define Power Platform DLP policies

1. Add custom prompt modifications

1. Conduct adversarial testing

1. Ensure BOMs can include subcomponents ? Configure an entity relationship

1. Report changes made by engineers ? Enable auditing

• Configure entity relationship
• Configure auditing

• Users logging in more frequently than average
• Logins occurring during non-business hours
• Patterns that may indicate compromised credentials or suspicious access

• Login History: Shows individual login attempts and basic details, but does not provide behavioral analysis or averages.
• Login Inspector: Used to troubleshoot or inspect a specific login attempt, rather than analyze organization-wide patterns.
• Login Report: May provide reportable login information, but it is not the specialized tool for forensic behavioral analysis.

• A. Modify CommunitiesSelfRegController

• C. Configure Communities registration to use a custom Visualforce page

• B. Modify the self-registration trigger — There is no standard “self-registration trigger” intended for this configuration. The registration logic is handled by the Visualforce controller.
• D. Configure a custom Apex controller — The standard approach is to modify the provided CommunitiesSelfRegController, rather than simply selecting a separate controller configuration.

python 
import azure.cognitiveservices.speech as speechsdk 


speech_config = speechsdk.SpeechConfig( 
    subscription="YOUR_KEY", 
    region="YOUR_REGION" 
)

Important Azure Speech SDK classes

• SpeechConfig

• AudioConfig

python 
  audio_config = speechsdk.audio.AudioConfig( 
      use_default_microphone=True 
  )

• SpeechRecognizer

python 
  recognizer = speechsdk.SpeechRecognizer( 
      speech_config=speech_config, 
      audio_config=audio_config 
  )

• SpeechSynthesizer

• AudioOutputConfig

 Bogotá, Colombia