Oracle · 1Z0-1159-26
Agentic AI for Oracle AI Database Professional (1Z0-1159-26) Practice Tests
Validates the ability to build AI agents, retrieval-augmented generation (RAG) pipelines, and semantic search solutions inside Oracle Database 23ai and Autonomous AI Database. Covers vector search, agentic AI design, and private agent factory deployments.
Exam Details
Practice Questions
400
≈ 6 practice exams
Duration
90 minutes
Passing Score
68%
Difficulty
ProfessionalLast Updated
Oct 2026
Topics Covered
Oracle AI Vector Search FoundationsApply Vector Distance and IndexingBuild Retrieval-Augmented Generation SolutionsDesign Select AI Agent SolutionsConfigure Agentic Interfaces and Tool AccessGovern Private Agent Factory Deployments
Agentic AI for Oracle AI Database Professional (1Z0-1159-26) Practice Exam Preparation
Use this 1Z0-1159-26 practice exam to prepare for Agentic AI for Oracle AI Database Professional (1Z0-1159-26) with realistic questions, detailed explanations, and focused study modes. The practice bank includes 400 questions for Oracle 1Z0-1159-26, so you can review the exam steadily instead of relying on one long cram session.
As you practice, pay extra attention to recurring topics such as Oracle AI Vector Search Foundations, Apply Vector Distance and Indexing, Build Retrieval-Augmented Generation Solutions, Design Select AI Agent Solutions, and Configure Agentic Interfaces and Tool Access. Start with short sessions to identify weak areas, then move into timed quizzes once your accuracy is consistent.
The explanations are written for candidates preparing for Oracle certification questions and reviewing how the concepts apply in practice. Use the free preview first, then unlock the full question bank when you are ready to build a complete study routine.
Sample Questions
5 sample questions with answers and explanations. The full bank has 400 questions, enough for 6 full-length practice exams.
Preview — answers shown1. Trey Research is creating an IVF vector index over the chunk table of its RAG store and is scripting a parameter sweep for NEIGHBOR PARTITIONS. Which two values would Oracle AI Database 26 accept for that parameter? (Select two!)
Multiple correct answersA-1
B10,000,000
C1
D10,000,001
E0
Explanation
The supplied Oracle CREATE VECTOR INDEX reference explicitly gives the IVF NEIGHBOR PARTITIONS range as >= 1 and <= 10000000. Thus 10,000,000 and 1 are accepted boundary values; 0, 10,000,001, and -1 are outside the range. Zero is allowed for the separate MIN_VECTORS_PER_PARTITION parameter, which does not establish its validity for NEIGHBOR PARTITIONS.
2. A capacity ticket at Nimbus Retail lists six candidate NEIGHBOR PARTITIONS values for new IVF vector indexes on Oracle AI Database 26ai, and the reviewer must approve only the values the database accepts. Which three values are valid? (Select three!)
Multiple correct answersA25,000,000
B10,000,001
C10,000,000
D1
E0
F250,000
Explanation
IVF NEIGHBOR PARTITIONS accepts integers from 1 through 10,000,000 inclusive, so both boundary values and anything between them, such as 250,000, are valid. Zero falls below the minimum, while 10,000,001 and 25,000,000 exceed the upper limit. This range counts clusters and should not be confused with the HNSW NEIGHBORS range of 2 through 2048, which counts graph connections per node.
3. A data scientist at Woodgrove Labs is validating query output by hand. Two 3-dimension vectors are stored as [1, 2, 3] and [4, 6, 3]. Which values will Oracle AI Vector Search return for EUCLIDEAN, EUCLIDEAN_SQUARED and MANHATTAN distance, in that order? (Select one!)
A5, 25, and 5
B5, 25, and 7
C7, 49, and 5
D25, 5, and 7
Explanation
The per-dimension differences are 3, 4 and 0. EUCLIDEAN is the square root of the summed squares, sqrt(9+16+0) = 5. EUCLIDEAN_SQUARED omits the square root, so it returns 25 and preserves the same ordering. MANHATTAN sums absolute differences, 3+4+0 = 7. Reversing the first two values swaps the squared and square-rooted forms; reporting 5 for Manhattan reuses the Euclidean result; and 7, 49, 5 computes Manhattan first and squares it, which mixes up the definitions.
4. Contoso Publishing plans an exact near-duplicate check over 30,000 article embeddings, using a self-join that compares each unordered pair once with a.article_id < b.article_id. Approximately how many VECTOR_DISTANCE evaluations will this run perform? (Select one!)
AAbout 15 million
BAbout 30 thousand
CAbout 900 million
DAbout 450 million
Explanation
An exact comparison of every unordered pair evaluates n(n-1)/2 distances: 30,000 times 29,999 divided by 2 is 449,985,000, about 450 million. Thirty thousand is the cost of a single exact query against all rows. Nine hundred million is n squared, which double counts pairs and self-comparisons, and 15 million matches no pairwise formula. Exact all-pairs cost grows quadratically, so at millions of rows you would run per-row approximate top-k queries against a vector index.
5. Bright Ledger's data team wants an MCP-enabled assistant to inspect an Autonomous AI Database, read schema details, create tables, and generate sample data from plain-language requests. Which of the official ADB MCP Server use-case categories does this describe? (Select one!)
AObservability
BReAct application development
CData modeling
DBusiness support
Explanation
Data modeling is the use case in which agents inspect the database, read schemas, create tables, and generate sample data through plain language. Observability instead covers querying metrics, AWR and ASH reports, and anomaly detection. Business support centers on natural-language SQL generation for ad-hoc analysis. ReAct application development is about building agentic applications on top of custom database tools rather than modeling the schema itself.
$17.99
One-time access to this exam
400 questions (6 practice exams' worth)
Unlimited timed exam simulations
Yours forever — no subscription
Detailed explanations
Free preview stays available