A context engineer is designing an MCP server for a Databricks agent that connects to a catalog of 30 tools. The server uses a progressive disclosure architecture to manage a strict token budget. At startup, the agent loads only the tool names and one-sentence descriptions. After receiving a user request to query a specific Unity Catalog table, the agent evaluates the catalog, identifies one tool as the likely match, and needs to move to the next stage of the architecture.
What should the agent do in the inspect layer?
A context engineer is building a retrieval-augmented generation agent on Databricks to answer questions about a large collection of legal contracts. Each contract is a structured PDF averaging 40 pages, with clearly defined sections such as “Definitions,” “Obligations,” and “Termination Clauses.” The embedding model in use has a 512-token context window. User queries are highly targeted, such as “What are the termination conditions in the indemnification clause?”, and require precise, clause-level retrieval rather than broad document summaries.
Which chunking strategy should the context engineer use?
A context engineer is investigating why a Genie space serving the sales operations team gives inconsistent answers to revenue questions. When business users ask the same QUESTION NO: across different sessions, for example, “What was Q3 EMEA revenue?”, Genie returns different figures. The space is configured over a Unity Catalog managed table of transaction-level data.
On reviewing the space, the engineer finds:
The underlying table has comprehensive table and column comments.
The space’s text instructions describe how revenue should be computed, including currency conversion against a daily FX-rate table, accrual-timing rules, and intercompany-elimination logic.
The space includes several sample questions covering quarterly and regional revenue patterns.
In the query history, Genie composes a new SQL statement for the revenue calculation in each session, and the FX-conversion and elimination steps differ from run to run.
Which change will most reliably stabilize the space’s answers to this team’s revenue questions?
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 examining an MLflow trace for a Databricks compliance agent mid-workflow on a quarterly regulatory review for a regional bank. The agent’s design pattern is to retrieve regulatory clauses, extract structured obligations with regional metadata and source citations, validate each obligation through the rules engine, and synthesize the audit memo from the structured intermediate plus the validation outcomes. The context window is nearing the compaction threshold, and compaction will fire before the workflow is complete.
The trace includes three executed spans and one queued span:
Span 1 (RETRIEVER): Retrieved 22 regulatory clause excerpts from the compliance vector index. Output: 22 documents (8,400 tokens) of raw clause text with metadata for source URI and section ID.
Span 2 (LLM/CHAT_MODEL): Produced a structured intermediate (410 tokens) containing three extracted compliance obligations, each annotated with a regional applicability tag describing how the bank’s state-level regulations modify the federal requirement and a source citation pointing back to the originating clause.
Span 3 (TOOL): Ran the three obligations through the rules engine and returned a structured pass/fail report (280 tokens) with one outcome entry per obligation. Each entry contains the obligation ID, the pass/fail value, and a brief explanation of the rule that determined the outcome.
Span 4 (LLM/CHAT_MODEL, queued): Will draft the compliance memo for the bank’s audit team.
Which content from the executed spans can the compactor safely summarize without affecting span 4’s execution?
A context engineer is developing a Databricks agent that autonomously manages a multi-day incident response workflow. The agent processes a continuous stream of structured log entries, tool call results, and short status updates. Recent messages tend to guide decisions, while earlier messages are often superseded by newer updates during the workflow. The engineer is considering whether to implement a hard-coded sliding-window trimming heuristic that drops messages older than a fixed turn count, or to invest in a semantic compaction approach that summarizes and condenses prior context before dropping it.
Which characteristic of this agent’s information relevance pattern justifies using the hard-coded sliding-window trimming heuristic instead of a semantic compaction approach?
A context engineer is building a retail agent with Agent Bricks Custom Agents. The agent has two registered tools: a SQL query tool that queries a live inventory database, and a vector search tool that searches a static product catalog vector index for product descriptions, specifications, and categories. Both tools are configured correctly, function properly, and return valid results when called.
At runtime, when users ask questions such as “Is this item in stock?”, the agent consistently calls the product catalog tool and provides a plausible availability answer without checking the live inventory database. The product catalog tool is not intended to answer live stock availability questions.
Which context engineering issue explains this behavior?
A context engineer is preparing to deploy a per-tenant Databricks agent for a SaaS retail analytics platform. Each platform customer is provisioned with their own Unity Catalog catalog, retail_ & lt;customer & gt;, and within each tenant catalog, data is organized by business domain rather than by origin.
The sales schema contains the customer’s POS transaction and returns tables, which the customer’s source systems write to throughout the day, alongside the daily_sales_by_store, hourly_revenue_by_category, and top_skus_weekly summary tables that a nightly Lakeflow Jobs pipeline produces by reading those POS tables. The inventory schema contains stock-level and inbound-shipment tables written continuously by ingestion, alongside demand_forecast_7d and reorder_suggestions tables produced by the SaaS’s ML pipelines. The customer schema contains customer master records and contact tables, all populated by ingestion.
The SaaS deploys one agent per customer through Model Serving, and each agent has its own service principal. Before any agent endpoint is registered, the governance requirements are that an agent for a given customer must be able to retrieve data only from that customer’s catalog and never from another customer’s catalog, and that within its own customer’s catalog, the agent must be able to retrieve only from tables populated directly by the customer’s ingestion pipelines, never from nightly summaries, forecasts, or other tables produced by SaaS pipelines.
Which governance approach should the context engineer implement for each tenant’s agent before registration?
A context engineer is evaluating a target-landscape research agent at a pharmaceutical R & D organization. Discovery scientists ask the agent to prepare a structured landscape memo about a proposed drug target: what has been studied, who else is in the space, at what clinical phase, and with what regulatory history.
The agent runs as a single agent on Agent Bricks Custom Agents. It retrieves from a Databricks AI Search index built over PubMed, ClinicalTrials.gov, internal program memos, and FDA review documents, and produces the memo in one pass within a single context window.
Narrow queries (one target, one indication) produce accurate memos. Broader landscape queries return with competitor programs missing that the agent had named earlier in its own reasoning, quoted evidence attached to the wrong study, and cross-document fields partially filled. The MLflow trace shows window utilization rising toward the cap before drafting begins. One agent runs throughout.
What change should the AI engineer make to the agent architecture’s boundaries?
A context engineer is building a compound AI system for a hospital network with Agent Bricks Custom Agents. The system answers clinician questions about a specific patient’s medication history. Patient records are stored in a Delta table that is continuously updated as new prescriptions, refills, and discontinuations are recorded, and clinicians must see the most current state of a patient’s record when they submit a query. HIPAA’s minimum necessary rule requires the system to retrieve only the requesting patient’s records instead of searching across the broader patient population. End-to-end latency per query must stay under two seconds.
Which retrieval strategy should the context engineer use?
|
PDF + Testing Engine
|
|---|
|
$42.5 |
|
Testing Engine
|
|---|
|
$32.5 |
|
PDF (Q&A)
|
|---|
|
$26.25 |
Databricks Free Exams |
|---|
|