Home/Certifications/Microsoft/Microsoft AI-103: Prompt Injection Defenses That Hold AI & Machine LearningUpdated October 11, 2026 · 7 min read An AI assistant can be asked to summarize a document that contains an instruction to ignore the user, expose secrets or call an unauthorized tool. The malicious text may arrive through an email, search result, web page or document chunk rather than a direct user request. This is prompt injection: lower-trust content attempts to cross the boundary into instructions that control the system. Microsoft Foundry Prompt Shields can help detect user and…

Home/Certifications/Microsoft/Microsoft AI-103: Private Networking for Azure AI AI & Machine LearningUpdated October 11, 2026 · 7 min read An AI application can use Microsoft Entra identities correctly and still expose services over a network path the organization does not intend to permit. A search index, model endpoint, storage account and agent tool may each have different public-access settings, private endpoint requirements and DNS behavior. Turning on private networking for one resource does not make the entire workflow private. Microsoft Foundry and Azure AI Search support network isolation options, but availability…

Home/Certifications/Microsoft/Microsoft AI-103: Online Evaluation of AI Systems AI & Machine LearningUpdated October 11, 2026 · 7 min read A model can pass a carefully curated prelaunch test set and still fail after real users introduce unfamiliar questions, document types or tool workflows. Online evaluation helps teams observe whether the running application produces useful, grounded and safe outcomes under actual conditions. But production interactions contain private content, variable authorization and delayed business results, so evaluation cannot mean sending every conversation to another model without controls. Microsoft Foundry supports evaluation of deployed…

Home/Certifications/Microsoft/Microsoft AI-103: Monitoring Model Drift in Azure ML AI & Machine LearningUpdated October 11, 2026 · 7 min read A predictive model can lose value even when its endpoint remains healthy and its response time barely changes. Customer behavior, source schemas, sensor calibration or business policy can shift away from the conditions represented in its training data. Azure Machine Learning model monitoring helps teams observe drift and data quality signals, but a statistical difference does not automatically mean the model is wrong. Engineers need to identify the reference population, interpret…

Home/Certifications/Microsoft/Microsoft AI-103: Designing Agent Memory with Boundaries AI & Machine LearningUpdated October 11, 2026 · 7 min read An assistant that forgets every conversation can frustrate users, but one that remembers too much can leak private information or act on outdated assumptions. The word memory often hides several different systems: a short conversation window, retrieved enterprise facts, durable user preferences, and state left by an unfinished tool workflow. Microsoft Foundry Agent Service includes managed memory capabilities, yet teams still need to decide what may persist, who can retrieve it, and…

Home/Certifications/Microsoft/Microsoft AI-103: MLOps and GenAIOps Together AI & Machine LearningUpdated October 11, 2026 · 7 min read An Azure application may combine a predictive model that scores risk with a generative assistant that explains the result and calls approved tools. These components change differently. The predictive model depends on training data, feature definitions, and measured prediction error; the assistant depends on prompts, retrieval, model versions, and tool permissions. Treating both as one vague “AI model” makes incidents hard to diagnose and releases hard to reproduce. MLOps and GenAIOps share foundations…

Home/Certifications/Microsoft/Microsoft AI-103: Reducing Azure AI Response Latency AI & Machine LearningUpdated October 11, 2026 · 7 min read A user experiences AI latency as the time between asking for help and getting useful progress, not as the raw speed of a model endpoint. Retrieval, tokenization, network calls, reasoning, tool execution, validation and final rendering all add delay. An application can stream text quickly while still spending a long time on the step that actually resolves the task, and aggressive shortcuts can reduce quality or security. Microsoft Foundry exposes model metrics…

Home/Certifications/Microsoft/Microsoft AI-103: Tuning Hybrid Search in Azure AI Search AI & Machine LearningUpdated October 11, 2026 · 7 min read A search query may contain both an exact product code and a description of what went wrong. Keyword ranking can recognize the code but overlook a paraphrased troubleshooting note; vector retrieval can find the concept yet return the wrong product revision. Azure AI Search hybrid queries combine these candidate sources, but their scores are not interchangeable and filtering can change the set of documents eligible for ranking. Good tuning starts…

Home/Certifications/Microsoft/Microsoft AI-103: Diagnosing Hallucinations in Azure AI AI & Machine LearningUpdated October 11, 2026 · 7 min read A plausible AI answer can be wrong in several different ways. The model may invent a citation, misread a retrieved policy exception, rely on out-of-date source material, or confidently answer a question the organization has never documented. Calling every failure a hallucination hides the cause and leads teams to adjust prompts when the real defect is search, permissions, or an unsafe workflow. Azure AI applications need a structured way to identify unsupported…

Home/Certifications/Microsoft/Microsoft AI-103: Grounding Azure AI in Enterprise Data AI & Machine LearningUpdated October 11, 2026 · 7 min read Enterprise data gives an AI assistant facts that a general-purpose model cannot reliably invent: the current contract clause, an approved internal procedure, or the version of a product a customer actually owns. Connecting that data to a model, however, can create new failure paths. A retrieved passage may be obsolete, belong to another tenant, or omit the exception that changes the answer. Grounding therefore means more than adding a search call…

Home/Certifications/Microsoft/Microsoft AI-103: Operating GenAIOps on Azure AI & Machine LearningUpdated October 11, 2026 · 7 min read Generative AI applications change along more dimensions than conventional web services. A release can modify a prompt, model deployment, retrieval index, content filter, or tool definition without touching the visible interface. Each change can alter accuracy, latency, cost, and authorization in different ways. GenAIOps brings those moving parts into a disciplined engineering process with evidence, versioning, evaluation, and rollback. Microsoft Foundry provides useful deployment and observability capabilities, but the operating model must still…

Home/Certifications/Microsoft/Microsoft AI-103: Moving Azure AI from Prototype to Production AI & Machine LearningUpdated October 11, 2026 · 7 min read A convincing AI prototype often works because one engineer knows which prompt to run, which documents to select, and how to recognize a bad answer. Production removes those hidden safeguards. Thousands of users introduce unfamiliar requests, access boundaries, burst traffic, and failures that the demo never faced. Microsoft Foundry provides services for deploying and observing AI applications, but it cannot decide which answers are acceptable or who is authorized to…

Home/Certifications/Microsoft/Microsoft AI-103: Event-Driven Azure AI Workflows AI & Machine LearningUpdated October 11, 2026 · 7 min read An AI feature often begins with an ordinary business event: a contract arrives, a support case changes status, or a new document is approved for search. Processing each event inside the original request makes the source system depend on model latency and third-party tools. Event-driven Azure architecture can decouple those concerns, but it also introduces retries, ordering questions, and temporary disagreement between systems. The right design distinguishes an event that reports a fact…

Home/Certifications/Microsoft/Microsoft AI-103: Durable AI Jobs with Queues AI & Machine LearningUpdated October 11, 2026 · 7 min read A document-processing assistant may take seconds to start and minutes to finish, especially when it extracts data, retrieves references, calls a model, and waits for a reviewer. Keeping the original HTTP request open during that chain creates fragile timeouts and encourages duplicate work when clients retry. A durable queue can separate accepting the job from performing it, but only if the workflow records ownership, retries safely, and gives users a reliable status….

Home/Certifications/Microsoft/Microsoft AI-103: Evaluation Datasets That Reveal Failure AI & Machine LearningUpdated October 11, 2026 · 7 min read An AI evaluation report can show a reassuring score even when the system repeatedly fails the situations that matter to users. An agent that handles straightforward questions may still expose a confidential document, execute the wrong tool, or miss an exception buried in a policy. The problem is often the dataset rather than the evaluator: convenient examples reward predictable demonstrations and ignore real operational variation. Microsoft Foundry supports structured evaluation datasets, but…