Searching CertSprout…
Searching CertSprout…
A little curiosity goes a long way
Search certifications, learning topics, career paths, public profiles and help across CertSprout.
Showing 20 results · Page 3 for “Microsoft”. More results are available below.
ACR-CS-PRO-M365 · Stakeholder engagement
Stakeholders have different interests, influence and information needs. Identify who is affected and who makes or supports each decision, then choose suitable engagement. Listen for concerns…
ACR-CT-PRO-M365 · Managed backup and recovery
Backup success and successful recovery are different observations. Test restoration, data usability and application dependencies against agreed objectives. RPO measures tolerable data loss, …
ACR-CT-PRO-M365 · MSP tenant boundaries
Check both customer scope and role permissions before performing a task. Parent policies may flow to child tenants or devices, while supported overrides change effective configuration. Inspe…
ACR-CT-PRO-M365 · Backup and recovery
Recovery point objective describes tolerable data loss measured in time; recovery time objective describes tolerable restoration time. Backup frequency influences the first, while restore sp…
ACR-CT-PRO-M365 · Storage concepts
Block storage presents volumes to an operating system, file storage organizes shared files and directories, and object storage addresses objects with metadata. The workload determines which …
AI-102 · Azure AI application concepts
An AI application may need language processing, document extraction, search, vision or generative output. Start with the input, expected output and acceptable error, then select a capability…
AI-102 · Generative AI systems
A generative model produces output based on its learned patterns and the context it receives. Retrieval can supply relevant documents, but the model may still misunderstand or invent details…
AI-102 · Responsible AI evaluation
Aggregate accuracy can hide poor performance for particular users or situations. Evaluate representative subgroups and realistic failure cases, and consider privacy, misuse and human oversig…
AI-103 · AI security boundaries
This concept review develops the reasoning needed for ai security boundaries. Start by identifying the relevant assets, permissions, evidence and trust boundaries. Distinguish what the obser…
AI-103 · Application and API design
An API contract defines accepted inputs, authorization, outputs and failure behavior. Clients need predictable responses and safe handling of timeouts and retries. Validate data at the bound…
AI-103 · Azure AI application concepts
An AI application may need language processing, document extraction, search, vision or generative output. Start with the input, expected output and acceptable error, then select a capability…
AI-103 · Generative AI systems
A generative model produces output based on its learned patterns and the context it receives. Retrieval can supply relevant documents, but the model may still misunderstand or invent details…
AI-200 · AI security boundaries
This concept review develops the reasoning needed for ai security boundaries. Start by identifying the relevant assets, permissions, evidence and trust boundaries. Distinguish what the obser…
AI-200 · Application and API design
An API contract defines accepted inputs, authorization, outputs and failure behavior. Clients need predictable responses and safe handling of timeouts and retries. Validate data at the bound…
AI-200 · Azure AI application concepts
An AI application may need language processing, document extraction, search, vision or generative output. Start with the input, expected output and acceptable error, then select a capability…
AI-200 · Generative AI systems
A generative model produces output based on its learned patterns and the context it receives. Retrieval can supply relevant documents, but the model may still misunderstand or invent details…
AI-300 · Data pipeline design
A data pipeline needs defined inputs, transformations, outputs and checkpoints. Failures can leave partial results, so decide how processing resumes without losing or duplicating records. Tr…
AI-300 · Databricks lakehouse concepts
A lakehouse workflow can separate raw ingestion from validated and business-ready datasets. Each layer should have a clear contract rather than merely a different folder name. Preserve enoug…
AI-300 · Machine-learning engineering
Data leakage occurs when training or model selection uses information that would not be available at prediction time. Split data in a way that respects time and related entities, and fit lea…
AI-300 · Responsible AI evaluation
Aggregate accuracy can hide poor performance for particular users or situations. Evaluate representative subgroups and realistic failure cases, and consider privacy, misuse and human oversig…
People search uses public names, handles and headlines. Private and unlisted profiles are excluded. Your goals are visible only to you in search.