Question 1

What is BDD (Behaviour-Driven Development)?

Accepted Answer

BDD is a collaborative practice where business, developers and testers agree on system behaviour through concrete examples before building it. Those examples are written in a shared language like Gherkin and can be automated as living documentation.

Question 2

What is a Three Amigos session?

Accepted Answer

It's a short meeting where a business person (usually the PO), a developer and a tester discuss a story and agree on concrete examples and acceptance criteria before development. It surfaces misunderstandings and edge cases early.

Question 3

What is the difference between BDD and TDD?

Accepted Answer

TDD is a developer practice of writing a failing unit test before the code, focused on design and correctness of units. BDD works at the behaviour level, uses business-readable examples, and involves the whole team.

Question 4

What is Cucumber and how does it work?

Accepted Answer

Cucumber is a tool that reads Gherkin feature files and runs each step by matching it to a step definition method written in code. It supports Java, JavaScript/TypeScript, Ruby and others; SpecFlow/Reqnroll and behave are similar tools for .NET and Python.

Question 5

What are the main Gherkin keywords?

Accepted Answer

Feature, Rule, Scenario (or Example), Given, When, Then, And, But, Background, Scenario Outline (or Scenario Template) with Examples, plus doc strings, data tables, tags and comments.

Question 6

What should go in Given, When and Then steps?

Accepted Answer

Given describes the starting state, When describes the single key action or event, and Then describes the observable outcome. A good scenario usually has one When, and Then checks outcomes the user or system can observe, not internals.

Question 7

What is the difference between declarative and imperative Gherkin?

Accepted Answer

Imperative scenarios list UI steps like clicking buttons and typing fields, while declarative scenarios describe intent, like 'When I log in as a premium user'. Declarative style is preferred because it reads like business rules and survives UI changes.

Question 8

How do you write a step definition in Cucumber?

Accepted Answer

A step definition is a method annotated with Given/When/Then and a Cucumber Expression or regex that matches the step text. Parameters like {int} or {string} are captured and passed as method arguments.

Question 9

What are Cucumber Expressions and how do they differ from regex?

Accepted Answer

Cucumber Expressions are a simpler alternative to regular expressions with built-in parameter types like {int}, {float}, {word} and {string}, optional text in parentheses and alternatives with a slash. Regex is still supported when you need more control.

Question 10

How do you create a custom parameter type?

Accepted Answer

In Cucumber-JVM, annotate a method with @ParameterType giving a regex; the method name becomes the parameter name and it converts the matched text into an object. Steps then use {name} in their expressions.

Question 11

What causes ambiguous or duplicate step definition errors?

Accepted Answer

An ambiguous error happens when one step text matches more than one step definition, and a duplicate error when the same expression is defined twice. Cucumber needs exactly one match per step, so I make expressions more specific or reuse the existing step.

Question 12

What is a Scenario Outline and when do you use it?

Accepted Answer

A Scenario Outline is a template run once per row of its Examples table, with <placeholders> replaced by column values. I use it when the same behaviour must be checked with several input-output combinations.

Question 13

How do you use data tables in Cucumber?

Accepted Answer

A data table passes structured data to a single step, unlike a Scenario Outline which runs the scenario many times. In the step definition I receive a DataTable and convert it to lists, maps or typed objects.

Question 14

Scenario Outline vs data table: what's the difference?

Accepted Answer

A Scenario Outline repeats the whole scenario for each Examples row, so each row is an independent test. A data table supplies multiple rows of data to one step within a single scenario run.

Question 15

What are doc strings in Gherkin?

Accepted Answer

Doc strings pass a multi-line block of text to a step, enclosed in triple quotes or triple backticks, optionally with a content type. They are handy for JSON payloads or email bodies.

Question 16

What is Background in a feature file and what are its pitfalls?

Accepted Answer

Background holds Given steps that run before every scenario in the feature or rule, avoiding repetition of common context. Pitfalls are making it long or including steps not all scenarios need, which hides important context.

Question 17

What hooks does Cucumber provide?

Accepted Answer

@Before and @After run around every scenario, @BeforeStep and @AfterStep around each step, and @BeforeAll and @AfterAll once per run. Hooks handle technical setup like starting browsers, screenshots and cleanup.

Question 18

How do you run a hook only for certain scenarios?

Accepted Answer

Use a tag expression on the hook, for example @Before("@db") to seed the database only for scenarios tagged @db. Tag expressions support and, or, not and parentheses.

Question 19

How do you use tags to select which scenarios run?

Accepted Answer

Tag features or scenarios with @smoke, @regression or ticket IDs, then run with a tag expression like "@smoke and not @wip" via the cucumber.filter.tags property, the runner configuration, or the --tags CLI option.

Question 20

How do you configure a Cucumber runner with JUnit 5?

Accepted Answer

With JUnit 5 you use the cucumber-junit-platform-engine and a suite class annotated with @Suite, @IncludeEngines("cucumber") and @SelectClasspathResource pointing to the features, plus configuration parameters for glue, plugins and tags. The older @CucumberOptions with @RunWith(Cucumber.class) is the JUnit 4 approach.

Question 21

How do you run Cucumber scenarios in parallel?

Accepted Answer

With TestNG, extend AbstractTestNGCucumberTests and override scenarios() with @DataProvider(parallel = true). With JUnit 5, set cucumber.execution.parallel.enabled=true in junit-platform.properties. Shared state like WebDriver must be per-thread.

Question 22

How do you share state between step definition classes?

Accepted Answer

Use dependency injection: with cucumber-picocontainer (or Spring/Guice), any class in the constructor of step classes is created once per scenario and shared between them. This gives scenario-scoped state without static variables.

Question 23

How do you structure a Cucumber + Selenium framework?

Accepted Answer

Feature files in src/test/resources/features, step definitions that call page objects, page objects that wrap Selenium, hooks for driver lifecycle and screenshots, and a runner class. A DI container shares the driver and context between steps.

Question 24

How would you use Cucumber with Playwright in TypeScript?

Accepted Answer

Use @cucumber/cucumber with a custom World that holds the Playwright browser, context and page, created in Before hooks and closed in After hooks. Alternatively, playwright-bdd generates Playwright tests from feature files so you keep the Playwright runner, fixtures and reporting.

Question 25

Can you use Cucumber for API testing?

Accepted Answer

Yes. Step definitions call an HTTP client such as REST Assured or Playwright's request context, store the response in the scenario context, and Then steps assert on status and body. Keep scenarios in business language, not raw HTTP details, unless the audience is technical.

Question 26

What reporting options does Cucumber offer?

Accepted Answer

Built-in plugins produce pretty console output, HTML, JSON, JUnit XML and message formats. Teams often add Allure or the masterthought cucumber-reporting library, and screenshots are attached via scenario.attach in hooks.

Question 27

What is dryRun in Cucumber?

Accepted Answer

dryRun checks that every step in the feature files has a matching step definition without actually executing them. It's a quick way to find undefined steps.

Question 28

What are common Cucumber anti-patterns?

Accepted Answer

Writing feature files after the code without collaboration, imperative UI-click steps, very long scenarios, scenarios depending on each other, too many Scenario Outline rows, and step definitions full of logic instead of delegating to page objects or services.

Question 29

When would you not use Cucumber?

Accepted Answer

When business stakeholders don't read or contribute to feature files, the Gherkin layer adds maintenance cost without benefit. For purely technical tests like unit, contract or low-level API checks, plain code tests are simpler.

Question 30

What makes a good BDD scenario?

Accepted Answer

It describes one behaviour, has a descriptive title, uses business language, contains only data relevant to the rule, and is independent of other scenarios. Ideally it's short, around three to seven steps.

Question 31

How would you improve this scenario: 'Given I open the browser, When I click login, And I type user, And I type password, And I click submit, Then I see home page'?

Accepted Answer

Make it declarative and outcome-focused: move browser setup to hooks, collapse the UI steps into one intent-level step, and assert a business outcome. For example: Given a registered customer, When they sign in with valid credentials, Then they see their account dashboard.

Question 32

What is living documentation?

Accepted Answer

Living documentation means the feature files describe the system's behaviour and are verified by automated runs, so they stay accurate as the system changes. Reports from each run show which behaviours currently work.

Question 33

How do you organise feature files in a large project?

Accepted Answer

Group features by business capability or domain, like checkout or payments, rather than by sprint or story number. Step definitions are organised by domain concept so they're reused across features.

Question 34

How do you rerun failed Cucumber scenarios?

Accepted Answer

Cucumber-JVM's rerun plugin writes failed scenario locations to a file, like rerun:target/rerun.txt, which a second runner reads with @target/rerun.txt as its feature path. Cucumber-JS has a built-in --retry option.

Question 35

How do you balance step reuse and readability?

Accepted Answer

I reuse steps when the wording stays natural for the business reader, but don't force awkward generic steps like 'I click {string}' everywhere. Reuse happens best at the code layer below steps, in page objects and helpers.

Question 36

How do acceptance criteria relate to Gherkin scenarios?

Accepted Answer

Acceptance criteria are the rules a story must satisfy, and Gherkin scenarios are concrete examples illustrating each rule. Usually each criterion has one or more scenarios covering the happy path and key edge cases.

Question 37

When do you use Background versus a Before hook?

Accepted Answer

Background is for business context the reader should see, like 'Given I am signed in as an admin'. Before hooks are for technical setup the reader doesn't care about, like launching the browser or clearing cookies.

Question 38

How do you configure Cucumber-JS with TypeScript?

Accepted Answer

Create a cucumber.js or cucumber.json config with paths to features, require or import for step files, and a TypeScript loader such as ts-node/register (requireModule) or tsx. Add formats for reports and parallel workers there too.

Question 39

How does BDD fit into an Agile sprint?

Accepted Answer

During refinement, Three Amigos turn stories into examples; during the sprint, developers and testers automate those scenarios alongside the code; and at the end, passing scenarios show the story is done. Scenarios then join the regression suite.

Question 40

What do undefined, pending and skipped step statuses mean?

Accepted Answer

Undefined means no step definition matches the text, pending means a definition exists but throws PendingException (or returns 'pending' in JS) because it's not implemented yet, and skipped means the step didn't run because an earlier step failed or was pending.

Interview Questions

01What is BDD (Behaviour-Driven Development)?BeginnerBDD Concepts

02What is a Three Amigos session?BeginnerBDD Concepts

03What is the difference between BDD and TDD?BeginnerBDD Concepts

04What is Cucumber and how does it work?BeginnerCucumber Basics

05What are the main Gherkin keywords?BeginnerGherkin

06What should go in Given, When and Then steps?BeginnerGherkin

07What is the difference between declarative and imperative Gherkin?IntermediateGherkin

08How do you write a step definition in Cucumber?BeginnerStep Definitions

09What are Cucumber Expressions and how do they differ from regex?IntermediateStep Definitions

10How do you create a custom parameter type?AdvancedStep Definitions

11What causes ambiguous or duplicate step definition errors?IntermediateStep Definitions

12What is a Scenario Outline and when do you use it?BeginnerData Handling

13How do you use data tables in Cucumber?IntermediateData Handling

14Scenario Outline vs data table: what's the difference?IntermediateData Handling

15What are doc strings in Gherkin?IntermediateData Handling