Database development test for hiring
Review how developers reason about relational data, schemas, concurrent sessions, and index choices in standard SQL.
- 10 āĻŽāĻŋāύāĻŋāĻ
- 15āĻāĻŋ āĻĒā§āϰāĻļā§āύ
- āĻŽāĻžāĻāĻžāϰāĻŋ
- āĻŦāĻšā§āύāĻŋāϰā§āĻŦāĻžāĻāύā§
- āĻāĻāϰā§āĻāĻŋ āĻāĻžāώāĻžāϝāĻŧ āĻĒā§āϰāĻļā§āύ
- āĻĢā§āύ āĻ āĻĄā§āϏā§āĻāĻāĻĒ
āϏāĻāĻā§āώāĻŋāĻĒā§āϤ āĻāϤā§āϤāϰ
The Database Development test is an intermediate, 10-minute check for developers who build or maintain relational databases. Its 15 multiple-choice questions use small tables, schemas, and SQL snippets to assess query results, schema design, transactions, and indexes in standard SQL, without vendor-specific features.
āĻāĻāĻŋ āĻā§ āĻĒāϰāĻŋāĻŽāĻžāĻĒ āĻāϰā§
Reasoning about relational data, query results, schema design, transactions and indexes in standard SQL.
Relational modeling and query results
Candidates read small tables and SQL snippets to decide what a query returns.
Schema design and normal forms
Candidates consider how a schema should be split and how normal forms apply.
Transactions and concurrency
Candidates reason about what happens when two sessions run at the same time.
Indexes and query performance
Candidates choose an index that helps a given query.
āĻĒā§āϰāĻļā§āύā§āϰ āϧāϰāύ
āĻŦāĻšā§āύāĻŋāϰā§āĻŦāĻžāĻāύā§
āĻĒā§āϰāĻžāϰā§āĻĨā§: āĻāĻāĻāĻŋ āϏāĻāĻā§āώāĻŋāĻĒā§āϤ āĻĒāϰāĻŋāϏā§āĻĨāĻŋāϤāĻŋ, āϏāĻžāϰāĻŖāĻŋ āĻŦāĻž āĻŦāĻŋāĻŦā§āϤāĻŋ āĻĒāĻĄāĻŧā§ āϏāĻ āĻŋāĻ āĻāϤā§āϤāϰāĻāĻŋ, āĻ āĻĨāĻŦāĻž āϏāĻŦ āϏāĻ āĻŋāĻ āĻāϤā§āϤāϰ āĻŦā§āĻā§ āύā§āϝāĻŧāĨ¤
āĻāĻĒāύāĻŋ āĻĻā§āĻā§āύ: āĻĒā§āϰāĻžāϰā§āĻĨā§ āĻŦāĻŋāώāϝāĻŧāĻāĻŋ āĻļā§āϧ⧠āĻāĻŋāύāϤ⧠āĻĒāĻžāϰā§āύ āĻāĻŋ āύāĻž āύāϝāĻŧ, āĻāĻžāĻā§ āϞāĻžāĻāĻžāύā§āϰ āĻŽāϤ⧠āϝāĻĨā§āώā§āĻ āĻāĻžāϞā§āĻāĻžāĻŦā§ āĻāĻžāύā§āύ āĻāĻŋ āύāĻžāĨ¤
10 āĻŽāĻŋāύāĻŋāĻā§ 15āĻāĻŋ āĻĒā§āϰāĻļā§āύ, āĻāĻāĻŦāĻžāϰ⧠āĻāĻāĻāĻŋ āĻāϰ⧠āĻĻā§āĻāĻžāύ⧠āĻšāϝāĻŧāĨ¤
āĻāĻ āĻĒāϰā§āĻā§āώāĻžāϰ āĻāĻĒāϝā§āĻā§ āĻĒāĻĻ
Developers who build and maintain relational databases: database developers, backend and full-stack engineers, data engineers, and people who manage or review database work.
What this test can show
Tracing a query by hand
Candidates work out what a query returns from small tables, showing whether they can predict a result, not only write SQL that looks right.
Two sessions at once
Transaction questions ask what happens when two sessions run at the same time, a skill that is hard to judge from a resume.
Splitting a schema
Schema questions show whether a candidate can see where a table should be split and apply normal forms to a concrete design.
No vendor advantage
Every question is written in standard SQL with no vendor-specific features, so a candidate who learned on one database product is not favored over another.
āĻĒā§āϰāϤāĻŋāĻāĻŋ āĻĒā§āϰāĻžāϰā§āĻĨā§āϰ āĻāύā§āϝ āĻāĻĒāύāĻŋ āϝāĻž āĻĒāĻžāĻŦā§āύ
- āĻĒā§āϰ⧠āϏāĻŽāϝāĻŧ āĻĒāϰā§āĻā§āώāĻžāϰ āĻā§āϝāĻžāĻŦā§āĻ āĻāĻŋāϞā§āύ
- āĻāĻĒāĻŋ-āĻĒā§āϏā§āĻ āĻāϰāĻž āĻāϤā§āϤāϰ āύā§āĻ
- āĻāĻāĻāĻžāύāĻž āĻāϤāĻŋāϤ⧠āĻāϤā§āϤāϰ āĻĻāĻŋāϝāĻŧā§āĻā§āύ
āϏā§āĻā§āϰ, āĻĒā§āϰā§āĻĢāĻžāĻāϞ āĻ āϏāϤāϤāĻžāϰ āϏāĻāĻā§āϤāĻā§āϞāĻŋ āϏāĻžāĻā§āώāĻžā§āĻāĻžāϰā§āϰ āĻĒāĻžāĻļāĻžāĻĒāĻžāĻļāĻŋ āĻāĻĒāύāĻžāϰ āĻĒāϰā§āϝāĻžāϞā§āĻāύāĻžāϰ āĻāύā§āϝ āĻĒā§āϰāĻŽāĻžāĻŖ, āĻāĻāύāĻ āϏā§āĻŦāϝāĻŧāĻāĻā§āϰāĻŋāϝāĻŧ āϏāĻŋāĻĻā§āϧāĻžāύā§āϤā§āϰ āĻāĻŋāϤā§āϤāĻŋ āύāϝāĻŧāĨ¤
āĻāĻ āĻā§āϏā§āĻāĻāĻŋ āĻŦāĻŋāύāĻžāĻŽā§āϞā§āϝ⧠āĻŦā§āϝāĻŦāĻšāĻžāϰ āĻāϰā§āύ
Test integrity signals
Each candidate receives a different set of questions from large question pools. Question order and answer options are shuffled, and the browser shows one question at a time. The server controls timing, so reloading returns the same question with only the remaining time. Question text cannot be copied or selected. HireValid records signals such as time away from the test tab, pastes into answers, and answers faster than a question can be read, then summarizes them in an Integrity Score and evidence timeline for employer review. It does not decide on its own. With candidate consent, an employer can add webcam photos at random moments. See the pre-employment testing guide for guidance on using test evidence.
āĻāĻŽāϰāĻž āϝā§āĻāĻžāĻŦā§ AI āĻŦā§āϝāĻŦāĻšāĻžāϰ āĻāϰ⧠āĻĒā§āϰāϤāĻžāϰāĻŖāĻž āĻ ā§āĻāĻžāĻ Âˇ āĻ āĻāĻŖā§āĻĄāϤāĻž-āϏāĻāĻā§āϰāĻžāύā§āϤ āϏāĻāĻā§āϤ
Use the score as one piece of evidence
Review the score alongside a job-relevant work sample and a structured interview. For a work sample, you could ask the candidate to reason through a small schema or SQL scenario similar to the skills covered here. The score is evidence for a person to review, not a decision on its own. Compare the coverage with SQL Coding: Querying and read how we build our tests when choosing an assessment.
Structured interview questions
Walk me through how you check what a query will return before you run it against real data.
Listen for a step-by-step way of tracing the SQL against the rows, rather than a definition of a query.
Tell me about a schema you split, or would split. What problem did the split solve, and what did it cost?
Listen for reasoning about the structure and its tradeoffs.
Describe a time two processes changed the same data at once. What happened, and what did you do about it?
Listen for attention to concurrent behavior and transactions, and how the fix was checked.
How do you decide whether a new index is worth adding for a slow query?
Listen for reasoning tied to the query and to what the index costs.
āϏā§āĻā§āϰāĻŋāĻ āĻ āĻŦā§āĻā§āĻāĻŽāĻžāϰā§āĻ āϏāĻŽā§āĻĒāϰā§āĻā§ āĻāĻžāύā§āύ
āύā§āϝāĻžāϝā§āϝāϤāĻž āĻ āĻĒā§āϰāϝāĻŧā§āĻāύā§āϝāĻŧ āϏā§āĻŦāĻŋāϧāĻž
Questions use neutral schemas such as orders, rooms, shipments and staff rosters, with names from many backgrounds, no money symbols and no local knowledge. All code is standard SQL, so no particular database product is favored, and behavior that differs between products is never the basis of an answer. The reading level is kept plain, and scratch paper is allowed for every candidate.
āϝā§āĻā§āϤāĻŋāϏāĻā§āĻāϤ āϏā§āĻŦāĻŋāϧāĻž āĻšāĻŋāϏā§āĻŦā§ āĻāĻĒāύāĻŋ āϝā§āĻā§āύ⧠āĻĒā§āϰāĻžāϰā§āĻĨā§āĻā§ āĻ āϤāĻŋāϰāĻŋāĻā§āϤ āϏāĻŽāϝāĻŧ āĻĻāĻŋāϤ⧠āĻĒāĻžāϰā§āύāĨ¤ āĻĢāϞāĻžāĻĢāϞ āĻāĻāĻāύ āĻŦā§āϝāĻā§āϤāĻŋāϰ āĻĒāϰā§āϝāĻžāϞā§āĻāύāĻžāϰ āĻāύā§āϝ āĻĒā§āϰāĻŽāĻžāĻŖ, āĻāĻāύāĻ āϏā§āĻŦāϝāĻŧāĻāĻā§āϰāĻŋāϝāĻŧ āϏāĻŋāĻĻā§āϧāĻžāύā§āϤ āύāϝāĻŧāĨ¤
Provisional: reliability is reported once 200 candidates have completed it.
āĻĒāϰā§āĻā§āώāĻžāϰ āĻŦāĻŋāĻŦāϰāĻŖ
A short, timed check of practical relational database skills for working developers. Candidates read small tables, schemas and SQL snippets and decide what a query returns, how a schema should be split, what happens when two sessions run at once, and which index helps a query. Every question is written in standard SQL with no vendor-specific features.
- āĻĒā§āϰāĻļā§āύāĻā§āϞāĻŋ āĻāĻāϰā§āĻāĻŋāϤā§āĨ¤
- āϝ⧠āϏāĻšāĻžāϝāĻŧāĻ āĻāĻĒāĻāϰāĻŖ āĻŦā§āϝāĻŦāĻšāĻžāϰ āĻāϰāĻž āϝāĻžāĻŦā§: āĻāϏāĻĄāĻŧāĻž āĻāϰāĻžāϰ āĻāĻžāĻāĻāĨ¤
- āĻĒā§āϰāϤāĻŋāĻāĻŋ āĻĒā§āϰāĻļā§āύ āĻā§āϰāĻŽāĻžāύā§āϏāĻžāϰ⧠āĻāĻāĻŦāĻžāϰ āĻāϰ⧠āĻāϤā§āϤāϰ āĻĻāĻŋāϤ⧠āĻšāϝāĻŧāĨ¤ āĻāĻā§āϰ āĻĒā§āϰāĻļā§āύ⧠āĻĢāĻŋāϰ⧠āϝāĻžāĻāϝāĻŧāĻžāϰ āϏā§āϝā§āĻ āύā§āĻāĨ¤
- āϏāĻŽāϝāĻŧāĻŦāĻĻā§āϧ āĻĒāϰā§āĻā§āώāĻžāϰ āĻāĻā§ 4 āĻŽāĻŋāύāĻŋāĻā§ 4āĻāĻŋ āĻ āύā§āĻļā§āϞāύā§āϰ āĻĒā§āϰāĻļā§āύāĨ¤
āĻĒā§āϰāĻžāϝāĻŧāĻļāĻ āĻāĻŋāĻā§āĻāĻžāϏāĻŋāϤ āĻĒā§āϰāĻļā§āύ
How does the Database Development test keep SQL questions neutral?+
It uses standard SQL and neutral schemas such as orders, rooms, shipments, and staff rosters. Names come from many backgrounds, questions use no money symbols or local knowledge, and behavior that differs between database products is never the basis of an answer.
Can candidates use scratch paper or try the Database Development practice set?+
Yes. Scratch paper is allowed for every candidate. The practice set has 4 questions and takes 4 minutes, so candidates see how the schemas and SQL snippets are shown before the timed test.
Which roles suit the intermediate Database Development test?+
It is intended for developers who build and maintain relational databases: database developers, backend and full-stack engineers, data engineers, and people who manage or review database work. Questions sit at an intermediate level, such as reading a schema, predicting a query result and choosing an index for a given query.
Can candidates go back or use a phone on the Database Development test?+
Candidates can take the test on a phone or a desktop. Each of the 15 questions is answered once, in order, with no going back, so a candidate commits to what a query returns or what a second session sees before moving on. Some questions have one correct answer and others ask for every option that applies.
āĻ āϏāĻŽā§āĻĒāϰā§āĻāĻŋāϤ āĻĒāϰā§āĻā§āώāĻž
āĻĄā§āĻāĻžāĻŦā§āϏ āĻŦā§āϝāĻŦāϏā§āĻĨāĻžāĻĒāύāĻž āĻ āĻĒā§āϰāĻļāĻžāϏāύ
10 āĻŽāĻŋāύāĻŋāĻ, āĻā§āĻŽāĻŋāĻāĻž-āύāĻŋāϰā§āĻĻāĻŋāώā§āĻ āĻĻāĻā§āώāϤāĻž
āĻĻā§āĻā§ āύāĻŋāύMicrosoft SQL Server
10 āĻŽāĻŋāύāĻŋāĻ, āϏāĻĢāĻāĻāϝāĻŧā§āϝāĻžāϰ-āϏāĻāĻā§āϰāĻžāύā§āϤ āĻĻāĻā§āώāϤāĻž
āĻĻā§āĻā§ āύāĻŋāύMySQL
10 āĻŽāĻŋāύāĻŋāĻ, āĻĒā§āϰā§āĻā§āϰāĻžāĻŽāĻŋāĻ āĻĻāĻā§āώāϤāĻž
āĻĻā§āĻā§ āύāĻŋāύāĻĄā§āĻā§āϞāĻĒāĻžāϰāĻĻā§āϰ āĻāύā§āϝ MariaDB
10 āĻŽāĻŋāύāĻŋāĻ, āϏāĻĢāĻāĻāϝāĻŧā§āϝāĻžāϰ-āϏāĻāĻā§āϰāĻžāύā§āϤ āĻĻāĻā§āώāϤāĻž
āĻĻā§āĻā§ āύāĻŋāύNoSQL āĻĄā§āĻāĻžāĻŦā§āϏ
10 āĻŽāĻŋāύāĻŋāĻ, āĻĒā§āϰā§āĻā§āϰāĻžāĻŽāĻŋāĻ āĻĻāĻā§āώāϤāĻž
āĻĻā§āĻā§ āύāĻŋāύPostgreSQL
10 āĻŽāĻŋāύāĻŋāĻ, āϏāĻĢāĻāĻāϝāĻŧā§āϝāĻžāϰ-āϏāĻāĻā§āϰāĻžāύā§āϤ āĻĻāĻā§āώāϤāĻž
āĻĻā§āĻā§ āύāĻŋāύāĻĒāϰāĻŦāϰā§āϤ⧠āύāĻŋāϝāĻŧā§āĻā§ āĻĄā§āĻāĻžāĻŦā§āϏ āĻĄā§āĻā§āϞāĻĒāĻŽā§āύā§āĻ āĻā§āϏā§āĻāĻāĻŋ āĻŦā§āϝāĻŦāĻšāĻžāϰ āĻāϰā§āύ
Free-āϏāĻš āĻĒā§āϰāϤāĻŋāĻāĻŋ āĻĒā§āϞā§āϝāĻžāύā§āĻ āϏāĻŽā§āĻĒā§āϰā§āĻŖ āĻā§āϏā§āĻ āϞāĻžāĻāĻŦā§āϰā§āϰāĻŋ āϰāϝāĻŧā§āĻā§āĨ¤