Databricks Certified Generative AI Engineer Associate Bundle

Not the strongest safeguard choice for this specific question. Version embeddings with source and model identifiers to prevent silently mixing incompatible representations. This action primarily relates to Embedding model swap and reindex compatibility: Changed embedding dimensions or model semantics typically require re-embedding and index consistency before comparing search quality. However, the described case — A support assistant retrieves a full 60-page manual and overwhelms the model prompt with unrelated sections. — is testing Chunk quality evaluation and boundary splitting. It does not directly establish the safeguard outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest safeguard choice for this specific question. Never compare metrics from non-equivalent data snapshots as if they were a paired model test. This action primarily relates to Experiment tracking and version reproducibility: MLflow experiments should retain prompt, model, dataset, parameters and evaluation outputs to support a defensible comparison. However, the described case — A support assistant retrieves a full 60-page manual and overwhelms the model prompt with unrelated sections. — is testing Chunk quality evaluation and boundary splitting. It does not directly establish the safeguard outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest safeguard choice for this specific question. Reassess tokenizer and context limits when switching embedding models or input languages. This action primarily relates to Embedding context window and chunk token distribution: The embedding encoder must accept the actual tokenized chunk length; silently truncating chunks can drop relevant suffix facts. However, the described case — A support assistant retrieves a full 60-page manual and overwhelms the model prompt with unrelated sections. — is testing Chunk quality evaluation and boundary splitting. It does not directly establish the safeguard outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest safeguard choice for this specific question. Do not treat request logging or gateway authentication as proof that underlying data access is authorized. This action primarily relates to AI Gateway rate limits, usage tables and inference telemetry: AI Gateway offers controlled access, supported usage tracking and rate limiting to protect shared endpoints and measure consumption. However, the described case — A team measures 'answer correctness' with a golden-answer comparator but its evaluation data includes only user questions. — is testing Ground-truth requirement for correctness versus reference-free judges. It does not directly establish the safeguard outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest safeguard choice for this specific question. Never assume diagnostics justify indefinite storage of unrestricted personal data or secrets. This action primarily relates to Inference logging and production trace privacy: Inference logs and tracing permit diagnosis of real issues but must be sampled, retained and redacted safely. However, the described case — A team measures 'answer correctness' with a golden-answer comparator but its evaluation data includes only user questions. — is testing Ground-truth requirement for correctness versus reference-free judges. It does not directly establish the safeguard outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest safeguard choice for this specific question. Keep monitoring schemas and retention governance versioned across model deployment changes. This action primarily relates to Inference tables and Agent Monitoring feedback loop: Inference tables and agent monitoring capture production request/response signals and traces for quality investigations. However, the described case — A team measures 'answer correctness' with a golden-answer comparator but its evaluation data includes only user questions. — is testing Ground-truth requirement for correctness versus reference-free judges. It does not directly establish the safeguard outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest diagnose choice for this specific question. Inspect heading structure, sentence breaks, table boundaries and parent document identifiers. This action primarily relates to Semantic boundaries versus fixed-window chunks: Headings, paragraphs and tables can require structure-aware splitting rather than arbitrary token windows. However, the described case — A website crawler indexes the same cookie banner thousands of times and retrieval returns banners above actual policy. — is testing Boilerplate and navigation filtering in source corpora. It does not directly establish the diagnose outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest diagnose choice for this specific question. Inspect labeled query/chunk relevance, false positive hits, top-k settings and evaluation coverage. This action primarily relates to Retrieval recall precision and ranking diagnostics: Measure hit recall, precision, MRR or nDCG on labeled query-to-relevant-chunk pairs before tuning retrieval. However, the described case — A website crawler indexes the same cookie banner thousands of times and retrieval returns banners above actual policy. — is testing Boilerplate and navigation filtering in source corpora. It does not directly establish the diagnose outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest diagnose choice for this specific question. Inspect MIME types, text-layer availability, encoding and parsing failures per source. This action primarily relates to Format-specific document extraction and OCR needs: Extract text with a parser suited to source format; image-only scans require OCR rather than HTML or plain-text parsing. However, the described case — A website crawler indexes the same cookie banner thousands of times and retrieval returns banners above actual policy. — is testing Boilerplate and navigation filtering in source corpora. It does not directly establish the diagnose outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest verify choice for this specific question. Inspect a full trace to confirm retrieved evidence and user intent reach the model and that citations point to real chunks. This action primarily relates to Retriever-generator chain component boundaries: A basic RAG chain separates user input, query transformation, retrieval, prompt assembly, model generation and output validation. However, the described case — A benefits intake app requires a JSON object with claim_id, decision and reason but the assistant returns Markdown paragraphs. — is testing Structured JSON schema prompt and format compliance. It does not directly establish the verify outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest verify choice for this specific question. Withhold one source during testing and confirm the chain degrades safely rather than inventing a missing record. This action primarily relates to Parallel retrieval and deterministic chain postprocessing: Combining structured lookup and semantic retrieval requires explicit routing, result joining, and downstream schema checks. However, the described case — A benefits intake app requires a JSON object with claim_id, decision and reason but the assistant returns Markdown paragraphs. — is testing Structured JSON schema prompt and format compliance. It does not directly establish the verify outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest verify choice for this specific question. Run adversarial quoted-text tests and compare output structure on representative and edge inputs. This action primarily relates to Instruction hierarchy and bounded prompt examples: Few-shot examples and explicit boundaries can clarify expected outputs, but embedded source documents should not become new system instructions. However, the described case — A benefits intake app requires a JSON object with claim_id, decision and reason but the assistant returns Markdown paragraphs. — is testing Structured JSON schema prompt and format compliance. It does not directly establish the verify outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest verify choice for this specific question. Simulate quality and latency threshold breaches and confirm responsible owners receive alerts. This action primarily relates to Production performance KPIs and incident thresholds: Monitoring should include answer correctness, retrieval coverage, safety, latency, drift and per-user harm indicators, not just availability. However, the described case — A team measures 'answer correctness' with a golden-answer comparator but its evaluation data includes only user questions. — is testing Ground-truth requirement for correctness versus reference-free judges. It does not directly establish the verify outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest verify choice for this specific question. Reproduce a failing request and verify a sanitized trace contains enough evidence to diagnose it. This action primarily relates to Inference logging and production trace privacy: Inference logs and tracing permit diagnosis of real issues but must be sampled, retained and redacted safely. However, the described case — A team measures 'answer correctness' with a golden-answer comparator but its evaluation data includes only user questions. — is testing Ground-truth requirement for correctness versus reference-free judges. It does not directly establish the verify outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest verify choice for this specific question. Simulate a traffic burst and confirm rate limiting protects other clients while usage reporting remains accurate. This action primarily relates to AI Gateway rate limits, usage tables and inference telemetry: AI Gateway offers controlled access, supported usage tracking and rate limiting to protect shared endpoints and measure consumption. However, the described case — A team measures 'answer correctness' with a golden-answer comparator but its evaluation data includes only user questions. — is testing Ground-truth requirement for correctness versus reference-free judges. It does not directly establish the verify outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest diagnose choice for this specific question. Inspect UC model names, signatures, version metadata, lineage and relevant ACLs. This action primarily relates to Unity Catalog registered model version and privileges: MLflow-logged models can be registered in Unity Catalog, with governed names, permissions and traceable version lineage. However, the described case — A policy chatbot creates vectors but never passes retrieved passages to the LLM, producing unsupported recommendations. — is testing Vector Search retrieval workflow and query integration. It does not directly establish the diagnose outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest diagnose choice for this specific question. Inspect prompt names, version history, aliases, tests and active referenced URI. This action primarily relates to MLflow Prompt Registry version aliases and promotion: Prompt Registry tracks immutable prompt versions with mutable aliases for staging and production, enabling change control and rollback. However, the described case — A policy chatbot creates vectors but never passes retrieved passages to the LLM, producing unsupported recommendations. — is testing Vector Search retrieval workflow and query integration. It does not directly establish the diagnose outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest diagnose choice for this specific question. Inspect logged model artifacts, environment requirements, signature and source index identifiers. This action primarily relates to RAG production bundle metadata and dependencies: A deployable RAG application needs tracked model code, retriever config, embedding model, dependency versions, signature and example input. However, the described case — A policy chatbot creates vectors but never passes retrieved passages to the LLM, producing unsupported recommendations. — is testing Vector Search retrieval workflow and query integration. It does not directly establish the diagnose outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest implement choice for this specific question. Use server-side ownership checks and transaction approval on the refund tool in addition to input screening. This action primarily relates to Prompt safety filtering and tool allowlists: Content filters and action allowlists are complementary; a moderation score is not a substitute for authorization of external effects. However, the described case — A support bot incorporates competitor FAQs with usage terms that forbid automated redistribution. — is testing Training and retrieval corpus license rights. It does not directly establish the implement outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest implement choice for this specific question. Redact or pseudonymize identity fields before vector indexing and enforce source access rules. This action primarily relates to Sensitive-field masking before embedding: Masking or tokenization can remove unnecessary PII from chunks before embedding and external model calls while retaining useful meaning. However, the described case — A support bot incorporates competitor FAQs with usage terms that forbid automated redistribution. — is testing Training and retrieval corpus license rights. It does not directly establish the implement outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest implement choice for this specific question. Isolate source text in retrieval context and use backend authorization for ticket actions. This action primarily relates to Input guardrails against prompt-injected source documents: A document that says 'ignore your rules' is untrusted data; a robust agent must enforce instruction and tool boundaries independently. However, the described case — A support bot incorporates competitor FAQs with usage terms that forbid automated redistribution. — is testing Training and retrieval corpus license rights. It does not directly establish the implement outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest verify choice for this specific question. Invoke the chain with a representative dictionary and check all intermediate variable mappings and final output. This action primarily relates to LangChain runnable composition and invocation: Composable runnable chains pass typed values between retrieval, prompt and model stages; each binding must preserve expected input keys. However, the described case — A retrieved wiki page tells the agent to email confidential documents to an external address. — is testing Prompt-injection defenses for retrieved context. It does not directly establish the verify outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest verify choice for this specific question. Audit answer-supportedness using labeled trace examples and a human review of fabricated citations. This action primarily relates to Hallucinated facts and unsupported citations: Fluent responses can invent facts and citations; faithfulness must be checked against supplied evidence, not writing style. However, the described case — A retrieved wiki page tells the agent to email confidential documents to an external address. — is testing Prompt-injection defenses for retrieved context. It does not directly establish the verify outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest verify choice for this specific question. Test sensitive-data leakage and harmful-output cases in both input and output stages. This action primarily relates to PII masking and output moderation as complementary guardrails: Input redaction and output checks serve distinct purposes; retain only necessary identifiers and validate actions separately. However, the described case — A retrieved wiki page tells the agent to email confidential documents to an external address. — is testing Prompt-injection defenses for retrieved context. It does not directly establish the verify outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest safeguard choice for this specific question. Keep component roles explicit so updates to a generation model do not silently alter retrieval vectors. This action primarily relates to Specialized model versus general LLM: Embedding, reranking, extraction and generative models solve different stages and should be evaluated for their actual role. However, the described case — A compliance summary app adds examples but stops honoring the one-line required answer format. — is testing Prompt changes evaluated against baseline. It does not directly establish the safeguard outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest safeguard choice for this specific question. Do not call a one-time benchmark a complete monitoring program for a deployed agent. This action primarily relates to Offline evaluation versus ongoing production monitoring: Offline test datasets catch regressions pre-release; production monitoring detects drift, real-user failures and evolving prompts. However, the described case — A compliance summary app adds examples but stops honoring the one-line required answer format. — is testing Prompt changes evaluated against baseline. It does not directly establish the safeguard outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest safeguard choice for this specific question. Do not tune chunk length on one successful demo; measure evidence recall and grounded answer performance. This action primarily relates to Chunk quality evaluation and boundary splitting: Retrieval failure can arise from overly small chunks that omit needed context or overly large chunks that dilute relevance. However, the described case — A compliance summary app adds examples but stops honoring the one-line required answer format. — is testing Prompt changes evaluated against baseline. It does not directly establish the safeguard outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest implement choice for this specific question. Combine keyword matching and vector similarity with model/tenant filters and calibrated score fusion. This action primarily relates to Hybrid sparse and dense retrieval with filters: Hybrid search combines lexical exact-match and semantic similarity; metadata filtering narrows authorized relevant content. However, the described case — An employee policy question needs a short deductible clause plus the adjacent exception paragraph. — is testing Parent-child retrieval and hierarchical structure. It does not directly establish the implement outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest implement choice for this specific question. Parse and chunk documents, write keyed chunk rows with provenance to a governed Delta table, then index them. This action primarily relates to Delta and Unity Catalog pipeline for chunk tables: A governed retrieval corpus stores chunk text, IDs, source lineage and metadata in managed Delta/Unity Catalog tables before indexing. However, the described case — An employee policy question needs a short deductible clause plus the adjacent exception paragraph. — is testing Parent-child retrieval and hierarchical structure. It does not directly establish the implement outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest implement choice for this specific question. Increase suitable chunk length and reduce unnecessary overlap while checking passage-level answer recall. This action primarily relates to Chunk size and overlap tradeoffs: Chunk size determines the number of vectors and coherence; overlap reduces boundary loss but increases index size and cost. However, the described case — An employee policy question needs a short deductible clause plus the adjacent exception paragraph. — is testing Parent-child retrieval and hierarchical structure. It does not directly establish the implement outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest implement choice for this specific question. Reject unauthorized tool requests through server-side permissions even if the model produces a valid tool call. This action primarily relates to Unsafe model output and jailbreak response diagnosis: User-provided instructions may attempt to bypass agent policies or cause data leakage; classification and tool boundaries need separate tests. However, the described case — A product assistant must answer only for the model and country selected in the user form. — is testing Query-field driven prompt and retrieval filters. It does not directly establish the implement outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest implement choice for this specific question. Provide a structured inventory lookup tool to the agent and ground the recommendation in its returned stock fields. This action primarily relates to Agent tool wrappers versus direct LLM responses: A tool-using agent needs explicit callable interfaces and typed tool results, whereas plain generation requires no action execution. However, the described case — A product assistant must answer only for the model and country selected in the user form. — is testing Query-field driven prompt and retrieval filters. It does not directly establish the implement outcome or constraint that the question requires; it may still be useful in a different stage or incident.

Not the strongest implement choice for this specific question. Route numerical KPI requests to the configured Genie Space then compose an answer from returned values. This action primarily relates to Multi-agent Genie Space tool routing and governed answers: Genie Spaces can provide governed natural-language access to structured business metrics; an agent should choose authorized Genie tools over guessing SQL facts. However, the described case — A product assistant must answer only for the model and country selected in the user form. — is testing Query-field driven prompt and retrieval filters. It does not directly establish the implement outcome or constraint that the question requires; it may still be useful in a different stage or incident.