Amazon Software Testing Interview Questions & Answers

Measuring speed and stability: load (expected users), stress (beyond limits), spike (sudden surge), endurance (long duration) and scalability testing.

Smoke testing checks that the critical features of a new build work (broad and shallow). Sanity testing checks that a specific fix or area works after a change (narrow and deep).

Valid login, invalid login, forgot password, locked account and session handling are the main scenarios to cover.

Valid login, wrong password, empty fields, case sensitivity, max length, SQL-injection strings, password masking, 'remember me', lock-out after N attempts and the forgot-password link.

Code is built and tested automatically on each change; tests are quality gates that stop faulty builds from moving forward.

Checking the application works across browsers, devices, operating systems, screen sizes and network conditions.

Re-check environment, data, browser and steps; check the logs; try the reporter's setup; add video and console logs; mark it 'cannot reproduce' only after discussing with the developer, and keep monitoring.

Statuses like To Do, In Progress, In Review, Ready for QA, Done and Reopened, with transitions that depend on the team's configuration.

Peer review for clarity, coverage of requirements, correct expected results, test data and traceability, then update and baseline them.

A defect that stops testing or release of a feature because no workaround exists, so it is highest priority.

A technique that tests all pairs of input values instead of every combination, greatly reducing test cases while catching most defects.

Mandatory fields, email format, password rules, duplicate email, terms checkbox, captcha, confirmation email and field length limits.

Assess impact and risk, discuss with product and dev, decide to fix, defer with a workaround or delay release, and document the decision.

Evaluating software to find defects and confirm it meets requirements. It reduces risk, protects users and saves the cost of fixing bugs after release.

The hardware, software, data and configuration where testing happens, set up to mirror production as closely as practical.

They let testers verify totals and counts per category, such as the number of orders per customer, and compare them with the UI or reports.

A language to query databases. Testers use it to verify data saved by the application, prepare test data and check backend results.

Prioritising tests by the likelihood and impact of failure, so critical areas get the most coverage when time is limited.

Severity is impact on the system; priority is urgency of the fix. A crash in a rare report is high severity, low priority; a spelling mistake in the logo is low severity, high priority.

Positive cases use valid inputs to confirm expected behaviour; negative cases use invalid inputs to confirm the system handles errors gracefully.

Add, remove, update quantity, price and tax calculation, coupon codes, stock limits, cart persistence after login and checkout navigation.

A scenario is a high-level thing to test (user logs in); test cases are the detailed steps and data that verify it.

Test execution rate, pass percentage, defect density, defect leakage, defect rejection ratio, requirement coverage and automation coverage.

A regular review where QA, development and product assess new bugs and decide severity, priority, owner and target release.

JIRA Query Language, used to search issues, for example project = ABC AND status = Open AND priority = High.

The number of defects per size of software, for example defects per thousand lines of code or per module, used to compare quality.

Success, failure and pending flows, timeouts, refunds, double-click payments, every method (UPI, cards, net banking), rounding, security (no card data stored) and reconciliation with bank reports.

Checking that people with disabilities can use the product, using screen readers, keyboard navigation, colour contrast and WCAG guidelines.

Learning, test design and execution at the same time, without scripted cases — time-boxed sessions with a charter (e.g. 'explore checkout with coupons'), noting bugs and ideas. Great for finding edge cases.

Unit, integration, system and acceptance testing, moving from individual components to the complete product in front of users.

Alpha is done in-house by internal teams before release; beta is done by a limited group of real users in their own environment.

A clear title, steps to reproduce, expected vs actual result, environment/build, severity and priority, and evidence (screenshot, video, logs).

A final document summarising what was tested, defects found and open, metrics, lessons learned and the recommendation for release.

Status code, response body and schema, headers, response time, error messages for invalid input, authentication (401/403) and that data is saved in the database.

A development model where each development phase has a matching testing phase, such as requirements with acceptance tests and design with integration tests.

Using experience to guess where defects likely hide, such as empty fields, special characters or unexpected sequences, and designing tests for them.

Testing how a system moves between states based on events, for example an order going from placed to paid to shipped to delivered.

A bug is behaviour that deviates from agreed requirements; an enhancement is a new or improved capability beyond the current requirements.

Send page, size, sort and filter parameters, then verify counts, order, boundaries (first, last, empty page) and invalid parameter handling.

A team meeting at the end of a sprint to discuss what went well, what did not and what to improve next.

Use mocks or stubs from the API contract such as Swagger, write test cases early and validate again when the real service is available.

Check SQL injection, brute-force lockout, password masking, session timeout, HTTPS, cookie flags, error messages not revealing users and password reset flow.

Check access with valid, expired and missing tokens, role-based permissions, error codes and that sensitive data is never exposed.

List conditions and actions in a table and cover every combination of conditions (rules) — ideal for business rules like loan approval: salary > X AND credit score > Y → approve.

Basic operations, division by zero, decimals, large numbers, order of operations, clear and backspace, repeated operations and memory keys.

A human error causes a defect in the code, and when that defect is executed it produces a failure visible to users.

Valid keyword, partial match, no results, special characters, very long text, empty search, case sensitivity, spelling errors, suggestions and speed of results.

Shift left with early reviews, improve test design, automate regression, use better test data, add monitoring and learn from escaped defects.

Designing tests from user flows described in use cases to verify complete real-world interactions.

INNER returns matching rows, LEFT keeps all left rows, RIGHT keeps all right rows and FULL keeps all rows from both tables.

Testing at the edges of input ranges, such as 17, 18 and 19 for an age limit of 18, since errors cluster at boundaries.

Allowed and disallowed formats, size limits, empty file, duplicate names, special-character names, cancel upload, slow network and multiple files.

TDD writes unit tests before code to drive design; BDD describes behaviour in business language scenarios that become automated acceptance tests.

Testing that is continuous and collaborative within each sprint, with testers working alongside developers and product owners from the start.

Keep tests modular, prioritise by risk, use exploratory testing, automate stable parts and keep close contact with product owners.

An epic is a large body of work; stories are user-facing slices of it; tasks are technical work items; bugs are defects.

An urgent patch released outside the normal schedule to fix a critical production defect, followed by regression testing and merge back to the main branch.

New → Assigned → Open → Fixed → Retest → Verified → Closed, with Reopened, Rejected, Deferred and Duplicate as side paths.

A grid mapping requirements to test cases and defects, proving every requirement is tested.

The extent of requirements or code exercised by tests. Measure requirement coverage with a traceability matrix and code coverage with tools.

Moves to the chosen floor, doors open and close, door sensor stops closing on obstruction, overload alarm, emergency button, behaviour in a power cut, max capacity and the floor display.

Verification: are we building the product right? (reviews, walkthroughs — no code run). Validation: are we building the right product? (actual testing against user needs).

200 OK, 201 Created, 204 No Content, 400 Bad Request, 401 Unauthorized, 403 Forbidden, 404 Not Found, 500 Internal Server Error.

Risk-based testing: cover critical user flows, recent changes and high-defect areas first. Clearly communicate the untested risks and get stakeholder sign-off.

Valid card and PIN, wrong PIN, three failed attempts, amount above balance, amount above daily limit, invalid denomination, machine out of cash, network failure and receipt printing.

Smoke tests run on every build for quick feedback; full regression runs less often, such as nightly or before release.

Evaluating how easy and pleasant a product is to use by observing real users complete tasks and noting confusion.

Internationalisation tests that the product can support many languages and formats; localisation tests that a specific locale is correct for text, currency, date and layout.

Defects found after release that testing should have caught; leakage % = defects in production / total defects x 100.

Testing services directly without the UI, checking status codes, responses, security and performance. It finds issues early and is faster than UI tests.

Static testing reviews documents and code without executing; dynamic testing executes the software to observe behaviour.

A model with many fast unit tests at the base, fewer API or integration tests in the middle and a small number of UI tests at the top.

Severity = impact on the system. Priority = urgency to fix. Logo misspelt on the homepage → low severity, high priority. Crash in a rarely used report → high severity, low priority.