Databricks Certified Context Engineer Associate Questions and Answers
A context engineer is building a customer support agent on Databricks. The agent’s session-initialization routine must populate a context object before turn 1 begins; the context object is then passed to every turn handler. The agent needs the user’s plan tier, region, and a few entitlement flags from the Lakebase customer_profile table in that context object. The values must be present before the first turn handler executes and may be referenced on any subsequent turn. The values do not change during a session, but could change across different sessions. The team wants to minimize per-turn Lakebase round-trips and the latency they add to turn responses.
Which retrieval approach should the engineer implement for this agent architecture?
A context engineer is debugging a Databricks agent that staffs an internal IT help desk. The agent is already connected to a retrieval pipeline over a live, continuously updated knowledge base, and relevant KB passages are retrieved and placed in its context on every request. Its system prompt has three parts: a 400-token persona preamble, a 600-token list of every supported software title, and a 200-token closing instruction block. None of these parts explains how to weigh retrieved passages against the model’s own training knowledge.
Observed behavior: the agent correctly identifies the user’s intent, and the correct KB passages are present in context. However, for version-specific questions, it answers from its training knowledge and recommends releases that the knowledge base marks as deprecated. The team will accept only a change that does not rewrite the full prompt and does not add or modify retrieval infrastructure.
Which system-prompt revision addresses the version-accuracy behavior?
A context engineer is debugging a Databricks customer support agent that manages complex billing escalations.
The system prompt instructs the agent to:
Strictly apply the company’s official refund policy.
Treat entries from the “Billing-Policies” tool as authoritative.
Treat FAQ snippets and prior chat history as non-binding guidance only.
In one multi-step trace, the engineer observes the following:
Step 1: The agent calls the “Billing-Policies” tool and retrieves the official refund policy, which clearly states that chargebacks filed more than 90 days after purchase are not eligible for refunds.
Step 2: The agent calls a “Customer-FAQ-Search” tool that returns a community-authored FAQ thread. A top-voted but outdated and unofficial answer incorrectly claims that chargebacks are fully refundable regardless of when they are filed, and the agent summarizes this in the running conversation as an established fact.
Step 3: In a later turn, the customer describes a chargeback dispute from 18 months ago. The agent reasons from its own earlier summary, approves a full refund, and makes no reference to the official policy section excluding cases older than 90 days, even though that policy text remains in the context.
Step 4: In subsequent similar cases, the agent repeatedly approves refunds for old chargebacks, citing its prior summary as justification each time.
Which context failure mode describes what is occurring in this trace?
A context engineer is debugging a Databricks agent that supports a multi-turn customer support application. The agent keeps a running summary of all conversation history across every user session in a single vector index and retrieves it on every turn through semantic search. Users report that the agent loses track of facts stated earlier in the same session, while unrelated details from other users’ past sessions appear unexpectedly. Latency has also increased as conversation volume grows.
Which storage configuration correctly resolves the memory type mismatch that is causing this degraded behavior?
A context engineer is building a regulatory compliance agent on Databricks that connects to an external compliance management platform through an MCP Service registered in Unity Catalog.
The MCP server exposes the following tools:
search_compliance_records: Retrieves existing compliance records without modifying them.
get_submission_status: Checks the status of previously submitted reports.
submit_regulatory_report: Sends a finalized regulatory report to an external authority.
delete_compliance_record: Permanently deletes an existing compliance record.
The agent serves compliance officers who are authorized to retrieve records and submit reports. However, organizational policy requires explicit human approval before every regulatory report submission.
The agent must continue executing read-only tools without interruption. It must also prevent unauthorized destructive operations and maintain an auditable record of tool invocations.
The engineering team has confirmed that the MCP tools function correctly. The remaining requirement is to enforce the approval and authorization rules at the platform governance layer rather than relying exclusively on instructions inside the agent ' s system prompt.
Which approach should the context engineer implement?
A context engineer manages a Genie space for a finance team. Users repeatedly ask the same shape of QUESTION NO: with different fiscal periods: “What is the total open pipeline by region for FY26 Q1?”, “...for FY26 Q2?” “...for the last quarter?”. The SQL pattern Genie generates is identical across these questions (same JOIN, GROUP BY, and aggregation, with only the fiscal-period filter changing), and the team requires that responses to these recurring questions be labeled Trusted in the chat UI. The engineer must register an asset that reuses the same vetted SQL pattern with the user-supplied fiscal period.
Which curation choice should the engineer make?
A context engineer is debugging a long-running data pipeline orchestration agent that runs multi-step workflows using registered Databricks tools such as run_job, query_catalog, and validate_schema. After compaction occurs at turn 40, the agent starts re-invoking tools it successfully called earlier in the session and ignores results that downstream steps depend on. Inspection of the compacted context shows that the summarization prompt was instructed to retain only the final state of each pipeline stage. The current compaction prompt reads: “Summarize the current pipeline state in a few sentences and include only what is necessary.”
Which revision to the compaction prompt addresses the coherence failure while keeping token cost low?
A context engineer is creating a Databricks agent to triage incoming customer support tickets for a B2B software platform. Each ticket averages 200 tokens.
For each ticket, the agent must:
(1) assign a severity rating of P0/P1/P2/P3 based on whether the issue is fully blocking, degrades core functionality, affects a non-critical feature, or is informational—a judgment that requires reading the ticket body rather than matching keywords;
(2) route the ticket to one of eight engineering teams based on the technical area implied by the ticket;
(3) generate a brief, context-aware acknowledgment reply.
The pipeline must process up to 500 tickets per minute with an end-to-end latency budget of under 2 seconds per ticket. The token budget per invocation is 2,048 tokens, and the operations team has flagged that inference costs must remain minimal at this throughput.
Which reasoning mode should the context engineer configure for the agent?
A context engineer is assessing a Databricks agent that coordinates a multi-week pharmaceutical supply chain compliance workflow across dozens of distribution centers. The agent uses a flat task decomposition strategy, dividing the entire workflow into a single-level list of atomic tasks that are dispatched and tracked independently. The workflow contains clusters of tightly coupled sub-tasks, such as per-facility inspection records that must all complete before a regional compliance certificate can be issued, and the agent repeatedly fails to coordinate these cluster-level completions correctly. Logs show that the agent loses track of intermediate group states, issues compliance certificates prematurely on incomplete facility data, and cannot recover gracefully when individual facility inspection tasks fail.
Which two changes should the context engineer make to resolve the observed failure?
Choose two
A context engineer is developing a Databricks agent for a retail inventory management system. The system prompt and tool descriptions consume 1,800 tokens of a 4,096-token context window, and the engineer reserves 1,000 tokens for the user turn and model response. The remaining tokens are used for few-shot examples, each costing approximately 250 tokens, so the engineer includes exactly five examples.
The agent uses three tools: inventory_lookup, reorder_check, and supplier_lead_time.
Six candidate examples are available:
Example 1: A warehouse manager asks whether a depleted SKU needs reordering; the agent calls inventory_lookup, confirms depletion, calls reorder_check, and returns a binary reorder decision.
Example 2: A procurement officer asks when a low-stock item can be replenished; the agent calls reorder_check, calls supplier_lead_time, and returns a replenishment recommendation.
Example 3: A store associate asks for the current stock count of a single product; the agent calls inventory_lookup and returns a scalar quantity.
Example 4: A store associate asks for the current stock count of a single product; the agent calls inventory_lookup and returns a structured status object containing the quantity and a restocking flag.
Example 5: A store associate asks whether a forecasted shortfall requires immediate action; the agent calls inventory_lookup, compares against minimums, calls reorder_check, and returns a binary reorder decision.
Example 6: A logistics coordinator asks how long a specific supplier takes to fulfill an order; the agent calls supplier_lead_time and returns a scalar lead-time value.
Which example should the engineer omit to preserve the token budget while maximizing behavioral coverage?
A context engineer is building a memory-augmented agent that stores user session memories in Lakebase as an OLTP Postgres table. Each memory record contains columns such as user_id, session_timestamp, preference_category, and confirmed_preference_value. At runtime, the agent must retrieve all confirmed preferences for a specific user that were recorded within the last 30 days. A Databricks AI Search index is created over conversation summary embeddings as part of a delta table synced from the Lakebase, providing an additional data source.
Which retrieval mechanism should the context engineer configure for this type of query?
A context engineer is developing a multi-agent research pipeline on Databricks. A coordinator agent divides an incoming scientific literature query into two parallel subtasks: a citation-retrieval agent and a methodology-analysis agent. Both agents independently receive only the original user query, produce their outputs, and return them to the coordinator, which then invokes a synthesis agent. The synthesis agent produces a final report that contradicts the citation-retrieval findings because it relies solely on the methodology-analysis output and has no visibility into retrieved citations.
Which two context propagation changes would prevent the conflicting outputs in the synthesis agent?
Choose two
A context engineer is developing a Databricks agent for a financial research platform that answers analyst queries using a vector search index over earnings reports. The agent uses a 180-token system prompt, a 250-token conversation history window, and a RAG pipeline that retrieves 8 document chunks at runtime. Each chunk contains 550 tokens, contributing 4,400 tokens of retrieved context. The model’s context window is 8,192 tokens, leaving just 3,362 tokens for the analyst query and model response. Profiling shows that the retrieved chunks consistently contain repetitive regulatory boilerplate and duplicated table headers that are rarely cited in the model’s final answers.
Which two changes should the context engineer make to reduce the token footprint of retrieved context without discarding relevant financial content?
Choose two
An AI engineer is auditing a Databricks compliance agent that has handled 65 conversation turns for a quarterly financial audit. The context window is at 92% capacity, and several audit findings remain open. The engineer wants to recover space by clearing specific raw tool outputs from the message history without compacting the conversation, and is reviewing four tool_result blocks. They also want to minimize the risk of losing information the agent still relies on.
The history contains:
A ledger reconciliation output at turn 7 returning 8,800 tokens of raw transaction rows from a read-only general ledger. The agent calculated aggregated taxable and non-taxable totals from these rows and passed them to the tax classification call at turn 10. The agent’s later replies and tool calls do not quote any underlying transaction rows.
A tax classification output at turn 10 returning 5,200 tokens of per-line bucket breakdowns across 312 transaction lines. The agent extracted the bucket totals and passed them to the compliance flag call at turn 13. The agent’s later replies and tool calls do not quote any per-line bucket assignments.
A compliance flag output at turn 13 returning 95 tokens with three flag codes and one-sentence explanations. The agent quotes specific flag codes and explanations from this output in turns 24, 38, and 61, incorporating them into the audit findings it is currently drafting.
A vendor master lookup output at turn 19 returning 3,400 tokens of supplier records and tax IDs from a read-only vendor master database. The agent quotes a vendor name and tax ID from this output in turns 41 and 57.
Which tool outputs should the engineer clear?
A context engineer is investigating a multi-agent pipeline on Databricks where a coordinating agent breaks down a document analysis task and dispatches subtasks to three specialized sub-agents. After deployment, the sub-agents produce outputs that are internally consistent but contradict one another and do not reflect earlier inferences made at the pipeline level. The sub-agents have access to the same retrieval index, the embedding model is verified as current, and individual sub-agent outputs are accurate when tested in isolation with full context supplied manually.
What is a reasonable root cause of the downstream failures?