Databricks Certified Generative AI Engineer Associate Bundle

Not the strongest implement choice for this specific question. Differentiate direct identifiers from permissible temporal fields and apply purpose-specific protection. This action primarily relates to Contextual masking versus destructive data removal: Masking must preserve information needed for the task, such as non-sensitive dates or amounts, and cannot replace access control. However, the described case — A retrieved internal wiki page embeds a fake system instruction to grant an attacker access to all support tickets. — is testing Input guardrails against prompt-injected source documents. 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. Review license and reuse conditions and use authorized sources or link without copying protected text. This action primarily relates to Training and retrieval corpus license rights: Public accessibility does not automatically authorize copying, embedding or using content in commercial RAG services. However, the described case — A retrieved internal wiki page embeds a fake system instruction to grant an attacker access to all support tickets. — is testing Input guardrails against prompt-injected source documents. 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. Mask card numbers and nonessential identity fields before ingestion and keep approved case references separately. 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 retrieved internal wiki page embeds a fake system instruction to grant an attacker access to all support tickets. — is testing Input guardrails against prompt-injected source documents. 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 safeguard choice for this specific question. Remove repeated footer noise while preserving meaningful classification metadata separately. This action primarily relates to Boilerplate and navigation filtering in source corpora: Duplicated headers, disclaimers and navigation menus contaminate embeddings and cause irrelevant hits. However, the described case — A manuals corpus produces 2.5 million vectors and exceeds the planned index capacity after 120-token chunks. — is testing Chunk size and overlap tradeoffs. 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. Separate missing knowledge from bad generation during incident triage and refresh upstream data responsibly. This action primarily relates to Authoritative source coverage for factual answers: A RAG model cannot answer accurately if the required authoritative record is missing or stale in its index. However, the described case — A manuals corpus produces 2.5 million vectors and exceeds the planned index capacity after 120-token chunks. — is testing Chunk size and overlap tradeoffs. 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 verify choice for this specific question. Reproduce representative task evaluation in the intended language and document limitations. This action primarily relates to Model cards and task-specific limitations: A model card records intended uses, supported language/context, safety limitations and evaluation scope relevant to selection. However, the described case — A helpdesk bot sends full payment card numbers to an external model for a simple FAQ answer. — is testing PII masking and output moderation as complementary guardrails. 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 repeatable attack cases and verify restricted content and destructive operations remain inaccessible. 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 helpdesk bot sends full payment card numbers to an external model for a simple FAQ answer. — is testing PII masking and output moderation as complementary guardrails. 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. 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 helpdesk bot sends full payment card numbers to an external model for a simple FAQ answer. — is testing PII masking and output moderation as complementary guardrails. 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. Retest adversarial sources after adding new file types and tool connectors. 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 financial invoice extractor masks every number, including legitimate tax amounts and invoice totals. — is testing Contextual masking versus destructive data removal. 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 grant the model blanket tool permission based only on a safe-looking input text score. 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 financial invoice extractor masks every number, including legitimate tax amounts and invoice totals. — is testing Contextual masking versus destructive data removal. 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. Maintain an auditable source-rights register and recheck rights when a corpus is refreshed. This action primarily relates to Training and retrieval corpus license rights: Public accessibility does not automatically authorize copying, embedding or using content in commercial RAG services. However, the described case — A financial invoice extractor masks every number, including legitimate tax amounts and invoice totals. — is testing Contextual masking versus destructive data removal. 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 use persuasive phrasing or a citation marker as evidence that a generated statement is true. 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 proposed LLM wins on general benchmark score but fails grounded FAQ questions more frequently than incumbent. — is testing Model evaluation with paired benchmarks. 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 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 proposed LLM wins on general benchmark score but fails grounded FAQ questions more frequently than incumbent. — is testing Model evaluation with paired benchmarks. 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. Revalidate model choice after volume or source-document shape changes materially. This action primarily relates to Model selection by quality latency and budget: Select the smallest suitable model that meets a task's accuracy and latency constraints under realistic workload, not marketing rank. However, the described case — A proposed LLM wins on general benchmark score but fails grounded FAQ questions more frequently than incumbent. — is testing Model evaluation with paired benchmarks. 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. Design a precise instruction on evidence and uncertainty, then compare against baseline using the same evaluation set. This action primarily relates to Prompt changes evaluated against baseline: A prompt improvement should be measured against a fixed baseline set, not accepted because a few outputs sound better. However, the described case — A team wants to deploy a new prompt based solely on improving live thumbs-up counts. — is testing Offline evaluation versus ongoing production monitoring. 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 cherry-pick average metrics that hide severe high-impact task failures. This action primarily relates to Model evaluation with paired benchmarks: Accuracy, faithfulness, relevance and refusal behavior require task-specific scoring, and a model switch needs common test conditions. However, the described case — A team wants to deploy a new prompt based solely on improving live thumbs-up counts. — is testing Offline evaluation versus ongoing production monitoring. 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. Extract model/country fields and apply them as verified metadata filters to approved retrieval. This action primarily relates to Query-field driven prompt and retrieval filters: User intent and structured fields such as account_id should drive authorized SQL lookups or retrieval filters, not be ignored in free text. However, the described case — A supply-chain assistant must call an inventory API before recommending an available replacement part. — is testing Agent tool wrappers versus direct LLM responses. 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. Enforce policy-layer refusals and prevent disclosure through prompt and output checks without relying on wording alone. 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 supply-chain assistant must call an inventory API before recommending an available replacement part. — is testing Agent tool wrappers versus direct LLM responses. 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. Evaluate capable long-context models on legal clauses and latency/cost acceptance thresholds. This action primarily relates to Model selection by quality latency and budget: Select the smallest suitable model that meets a task's accuracy and latency constraints under realistic workload, not marketing rank. However, the described case — A supply-chain assistant must call an inventory API before recommending an available replacement part. — is testing Agent tool wrappers versus direct LLM responses. 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. Use access-controlled persistent storage with retention and sensitive-field minimization. This action primarily relates to Persistent conversation memory with authorized datastore: Transient prompt context differs from persistent user state; durable memory needs explicit storage, identity scope and retention. However, the described case — A RAG endpoint should be invoked only by the internal support group, but every workspace member can call it. — is testing Model Serving endpoint permissions and caller identity. 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. Register an approved external MCP service with secret-backed credentials and least-privilege access. This action primarily relates to Managed versus external versus custom MCP server integration: MCP standardizes tool discovery; managed servers reduce hosting, external services need credentials and custom servers wrap bespoke actions. However, the described case — A RAG endpoint should be invoked only by the internal support group, but every workspace member can call it. — is testing Model Serving endpoint permissions and caller identity. 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. Build a CI gate for prompt schema, retrieval relevance, agent tools and controlled promotion across environments. This action primarily relates to CI/CD pipelines for retrieval and agent components: Production promotions should test source index, prompt, model packaging and tool behavior independently before release. However, the described case — A RAG endpoint should be invoked only by the internal support group, but every workspace member can call it. — is testing Model Serving endpoint permissions and caller identity. 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. Translate SLA into retrieval latency, time-to-first-token, maximum context and quality floor for candidate designs. This action primarily relates to Latency, cost and retrieval targets as design requirements: A valid pipeline design has measurable quality, latency, token and throughput constraints before model selection. However, the described case — An internal service team wants a citation-backed chat over equipment procedures without writing a custom agent framework. — is testing Agent Bricks Knowledge Assistant suitability. 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. Call a structured order-status tool for transaction fields and retrieve policy text before composing a grounded response. 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 — An internal service team wants a citation-backed chat over equipment procedures without writing a custom agent framework. — is testing Agent Bricks Knowledge Assistant suitability. 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. Separate low-trust quoted input from higher-priority task instructions and include an example that preserves the schema. 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 — An internal service team wants a citation-backed chat over equipment procedures without writing a custom agent framework. — is testing Agent Bricks Knowledge Assistant suitability. 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. Enable approved inference tracing/logging with request IDs, redaction and retrieval metadata. 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 finance team cannot attribute escalating LLM spend to teams because requests have no workload identity. — is testing AI Gateway rate limits, usage tables and inference telemetry. 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. Define scoring anchors, calibrate experts on disputed samples and use aligned feedback as reliable evaluation labels. This action primarily relates to Human SME feedback rubric and reviewer calibration: Domain experts should use a shared rubric, calibrate disagreements and maintain adjudicated examples to improve model evaluation. However, the described case — A finance team cannot attribute escalating LLM spend to teams because requests have no workload identity. — is testing AI Gateway rate limits, usage tables and inference telemetry. 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. Configure approved inference tables and trace metadata to correlate outputs with deployed model/index versions. 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 finance team cannot attribute escalating LLM spend to teams because requests have no workload identity. — is testing AI Gateway rate limits, usage tables and inference telemetry. 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 safeguard choice for this specific question. Use an OCR-capable extraction step, evaluate recognition quality and retain source page provenance. 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 — Hybrid search retrieves the right troubleshooting note at rank 18 yet generation uses only the first five hits. — is testing Cross-encoder reranking after candidate retrieval. 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. Retain source offsets and section IDs so citations remain traceable after rechunking. 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 — Hybrid search retrieves the right troubleshooting note at rank 18 yet generation uses only the first five hits. — is testing Cross-encoder reranking after candidate retrieval. 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.