Databricks Certified Generative AI Engineer Associate Bundle

Not the strongest safeguard choice for this specific question. Keep long-lived credentials out of clients and distinguish shared app authorization from user impersonation. This action primarily relates to Databricks Apps user-facing agent interface and OAuth identities: Databricks Apps supports authenticated interactive frontends; app credentials and user credentials have different scopes. However, the described case — A legacy SQL Classic warehouse cannot run a new `ai_query()` workload and engineers assume bad model credentials. — is testing SQL ai_query for governed batch inference. 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. Pin compatible dependencies and rebuild/evaluate the serving artifact before promoting a new environment. This action primarily relates to Deployed model dependency drift and environment compatibility: Models logged in experiments can fail in serving if Python dependencies or inference signatures differ between build and target. However, the described case — A legacy SQL Classic warehouse cannot run a new `ai_query()` workload and engineers assume bad model credentials. — is testing SQL ai_query for governed batch inference. 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. 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 deployed claims assistant has 99.99% uptime but increasing incorrect eligibility advice. — is testing Production performance KPIs and incident thresholds. 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. Limit retrieved context by relevance and compare answer faithfulness before switching to a smaller endpoint. This action primarily relates to Databricks usage and token cost controls: Reduce LLM spend through token budgets, caching, smaller appropriate models and AI Gateway quotas while preserving acceptance quality. However, the described case — A deployed claims assistant has 99.99% uptime but increasing incorrect eligibility advice. — is testing Production performance KPIs and incident thresholds. 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. Apply supported AI Gateway endpoint quotas or rate limits by workload and inspect usage traces. 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 deployed claims assistant has 99.99% uptime but increasing incorrect eligibility advice. — is testing Production performance KPIs and incident thresholds. 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. Do not treat a model hub popularity score as certification of deployment fitness. 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 team switches from a 768-dimensional embedding model to a 1024-dimensional model while keeping the same vector index. — is testing Embedding model swap and reindex compatibility. 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 team switches from a 768-dimensional embedding model to a 1024-dimensional model while keeping the same vector index. — is testing Embedding model swap and reindex compatibility. 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 switches from a 768-dimensional embedding model to a 1024-dimensional model while keeping the same vector index. — is testing Embedding model swap and reindex compatibility. 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 row provenance and reject extraction transformations that detach numerical values from headers. This action primarily relates to Table-aware chunking and metadata-rich retrieval: Tabular rows need headers, units and column semantics preserved when transformed into retrieval documents. However, the described case — A team embeds PDFs directly into an unmanaged in-memory list, losing source versions and permissions. — is testing Delta and Unity Catalog pipeline for chunk tables. 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 team embeds PDFs directly into an unmanaged in-memory list, losing source versions and permissions. — is testing Delta and Unity Catalog pipeline for chunk tables. 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. Select chunk granularity by measured retrieval results, not solely by minimizing index records. 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 — A team embeds PDFs directly into an unmanaged in-memory list, losing source versions and permissions. — is testing Delta and Unity Catalog pipeline for chunk tables. 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 package lockfile, model signature, serving logs and API contract differences. This action primarily relates to Deployed model dependency drift and environment compatibility: Models logged in experiments can fail in serving if Python dependencies or inference signatures differ between build and target. However, the described case — A LangChain pipeline returns an SDK message object while its API gateway expects a plain answer string. — is testing Simple prompt-retrieval-generation chain assembly. 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 predict entrypoint, model artifacts, dependency environment, inputs and output serialization. This action primarily relates to MLflow pyfunc packaging for pre- and post-processing: A custom MLflow pyfunc model wraps request parsing, retriever/model invocation and output formatting inside a reproducible predict interface. However, the described case — A LangChain pipeline returns an SDK message object while its API gateway expects a plain answer string. — is testing Simple prompt-retrieval-generation chain assembly. 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 query embedding provenance, index schema, similarity results and selected context window. This action primarily relates to Vector Search retrieval workflow and query integration: Semantic retrieval uses an embedding model to query a compatible index, returns ranked passages and metadata for grounded generation. However, the described case — A LangChain pipeline returns an SDK message object while its API gateway expects a plain answer string. — is testing Simple prompt-retrieval-generation chain assembly. 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. Inject the hostile PDF and verify no secret leaves the backend and no unauthorized tool is invoked. 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 — An imported policy library contains a deliberately altered obsolete security procedure mixed into current instructions. — is testing Toxic, inaccurate and adversarial corpus cleaning. 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 malicious requests with valid JSON tool calls and confirm the tool rejects unauthorized actions. 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 — An imported policy library contains a deliberately altered obsolete security procedure mixed into current instructions. — is testing Toxic, inaccurate and adversarial corpus cleaning. 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. Compare usefulness and privacy leakage metrics on held-out documents before and after selective masking. 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 — An imported policy library contains a deliberately altered obsolete security procedure mixed into current instructions. — is testing Toxic, inaccurate and adversarial corpus cleaning. 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 gateway policies, invocation identity, consumption tables, endpoint errors and supported feature availability. 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 travel agent meets its average response time but p99 exceeds 30 seconds during peak traffic. — is testing Production performance KPIs and incident thresholds. 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 inference table availability, payload schemas, trace IDs and user-feedback mapping. 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 travel agent meets its average response time but p99 exceeds 30 seconds during peak traffic. — is testing Production performance KPIs and incident thresholds. 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 each judge's required inputs, reference fields, rubric and expected failure behavior. This action primarily relates to Ground-truth requirement for correctness versus reference-free judges: A correctness comparison often needs an expected answer; other judges such as format or safety can operate without a reference. However, the described case — A travel agent meets its average response time but p99 exceeds 30 seconds during peak traffic. — is testing Production performance KPIs and incident thresholds. 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. Resolve the authorized production alias at runtime using a controlled refresh strategy. 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 claims RAG prototype uses notebook variables and cannot run when served outside the developer session. — is testing MLflow pyfunc packaging for pre- and post-processing. 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. Move to a supported SQL warehouse/runtime and verify endpoint region/permission prerequisites. This action primarily relates to SQL ai_query for governed batch inference: `ai_query()` invokes an available serving endpoint from SQL; suitability depends on supported runtimes, endpoint and workload. However, the described case — A claims RAG prototype uses notebook variables and cannot run when served outside the developer session. — is testing MLflow pyfunc packaging for pre- and post-processing. 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 claims RAG prototype uses notebook variables and cannot run when served outside the developer session. — is testing MLflow pyfunc packaging for pre- and post-processing. 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 diagnose choice for this specific question. Inspect the business decision, data owners, actor permissions, known inputs and success criteria. This action primarily relates to Translate use-case requirements into pipeline inputs and outputs: Business acceptance criteria must map to concrete source inputs, generated fields, correctness measures and permission constraints. However, the described case — A CRM agent repeats a create-ticket tool call after a timeout, producing duplicate complaints. — is testing Agent tool schemas, least privilege and retries. 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 whether the task is grounded document Q&A, extraction, tool coordination or cross-agent orchestration. This action primarily relates to Agent Bricks Knowledge Assistant suitability: A governed knowledge assistant grounds conversational responses in curated documents and access-controlled retrieval. However, the described case — A CRM agent repeats a create-ticket tool call after a timeout, producing duplicate complaints. — is testing Agent tool schemas, least privilege and retries. 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 request volume, concurrency, response deadline, token budget and evaluation quality targets. 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 — A CRM agent repeats a create-ticket tool call after a timeout, producing duplicate complaints. — is testing Agent tool schemas, least privilege and retries. 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 safeguard choice for this specific question. Do not present nearest neighbors as verified answers without evidence validation and permission checks. This action primarily relates to Vector Search retrieval workflow and query integration: Semantic retrieval uses an embedding model to query a compatible index, returns ranked passages and metadata for grounded generation. However, the described case — A team promotes a new prompt without rerunning citation accuracy or tool-permission regression cases. — is testing CI/CD pipelines for retrieval and agent components. 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. Enforce reproducible environment manifests and API compatibility checks in CI/CD. This action primarily relates to Deployed model dependency drift and environment compatibility: Models logged in experiments can fail in serving if Python dependencies or inference signatures differ between build and target. However, the described case — A team promotes a new prompt without rerunning citation accuracy or tool-permission regression cases. — is testing CI/CD pipelines for retrieval and agent components. 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. Treat index freshness as a monitored dependency of source-document correctness. This action primarily relates to Databricks Vector Search index from governed Delta data: A vector index serves embeddings of chunked records maintained in a Unity Catalog governed Delta source table. However, the described case — A team promotes a new prompt without rerunning citation accuracy or tool-permission regression cases. — is testing CI/CD pipelines for retrieval and agent components. 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.