Software Testing Interview Questions for Senior Specialists (₹15–20 LPA)
40 questions with answers for senior specialists — 31 medium, 9 hard.
What ₹15–20 LPA interviews expect
Deep expertise in a specialisation Architecture-level reasoning Metrics and business impact Leading small initiatives
About ₹15–20 LPA Software Testing roles
Roles in the ₹15–20 LPA band are typically for senior specialists. Interviewers check whether you can already do the work expected at this level, so questions here are weighted towards Performance and Security Basics (7), Testing Types (5), Mobile Testing (5), Test Case Design (4).
Salary bands are indicative only. The same skills can pay differently across cities, company types and negotiation.
Solved in this bandStart practicing
3 QUESTIONS FREESign up to unlock the full question bank and track your progress.Unlock every question and answer, then continue practising company-wise and package-wise.
A flaky test passes and fails on the same code without any change. Common causes are timing problems such as fixed sleeps or missing waits, tests depending on each other or on shared data, unstable locators, animations, third-party services, test order in parallel runs and environment issues like slow servers. Flaky tests destroy trust in the suite, so teams quarantine them, find the root cause and fix it rather than adding retries forever.
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.
Backward compatibility testing checks that a new version still works with older versions of things it depends on or that depend on it. Examples: a new app version opens files and data created by the old one, a new API version still serves older mobile app versions that users have not updated, and an upgraded database keeps old records readable. It prevents breaking existing users when only some parts of the system are upgraded.
06
How do you link stories, test cases and bugs in Jira for traceability?
Create test cases as linked issues or in a test management add-on and link each one to its story with a tests relationship. When a test fails, create the bug from the test execution or link it to both the test and the story with is blocked by or relates to. Use the same fix version for all. A story then shows its tests and open bugs, which makes sign-off and the traceability report easy.
A performance baseline is a set of measurements of the current system under a known load, such as response times, throughput, error rate and resource usage. Every new build or release is compared with it to spot regressions. For example, if the checkout API had p95 of 800 milliseconds at 200 users last release and now shows 1.5 seconds, the change needs investigation. A benchmark compares against an industry or competitor standard.
08
How do you run a moderated usability test session?
Define goals and three to five realistic tasks, such as 'Book a train ticket for next Friday'. Recruit participants who match real users. In the session, explain that you are testing the product, not them, ask them to think aloud, and avoid hinting even when they struggle. Record the screen with consent, note where they hesitate or fail, and finish with a short questionnaire. Then group findings by severity and frequency.
09
How do you test biometric login with fingerprint or face unlock?
Test enabling biometrics in app settings after a password login. Verify login with a registered fingerprint or face, and failure with an unregistered one, falling back to PIN or password after the allowed attempts. Remove or add a fingerprint in device settings and check the app asks for re-authentication. Test devices without biometric hardware, disabled biometrics and app behaviour after an OS update or reinstall.
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.
11
What is the difference between a defect and a production incident?
A defect is a flaw in the software that makes it behave differently from the requirement, whether found in testing or production. A production incident is an event that disrupts or degrades service for users, such as an outage or a failed payment flow. An incident may be caused by a defect, but also by infrastructure, configuration or third-party failures. Incidents are handled by an incident process first, then linked to defects for the fix.
12
Your test repository has many duplicate and outdated test cases. How do you clean it up?
Export the cases and group them by module and requirement. Mark duplicates, merge them into one well-written case, and retire cases for removed features. Check which cases have not been executed or failed in months and confirm whether they are still valid. Link the remaining cases to requirements so gaps are visible. Do it module by module with peer review, and set a rule that every story update also updates its test cases.
The pesticide paradox says that if you run the same tests again and again, they stop finding new defects, just as pests become resistant to the same pesticide. To counter it, regularly review and update test cases, add new scenarios from recent production bugs, vary test data, and mix scripted runs with exploratory sessions. Rotating testers across modules also brings fresh eyes and new ideas.
15
How do you plan testing for a feature that depends on a third-party service?
Identify the integration points and the service's sandbox, limits and test data. Plan to test with a mock or stub early for our logic and error handling, then against the real sandbox for integration. Include failure scenarios: timeout, slow responses, invalid data and the service being down. Add the dependency to the risk list, with contacts and a fallback plan, and confirm production credentials and monitoring before go-live.
16
Tell me about a time you designed tests for a feature with unclear requirements.
Model answer: Situation: I got a story for 'flexible EMI options' with only one line of description. Task: test it within the sprint. Action: I listed my assumptions on tenure, interest and rounding, reviewed them with the product owner in a 15-minute call, and wrote test cases only after we agreed. Result: three rules were changed before coding finished, and the feature passed UAT without rework.
17
Describe a time an automated test caught a regression that manual testing would have missed.
Situation: a small tax rounding change was made late on a Friday. Task: no manual regression was planned for pricing. Action: our nightly API suite failed one data-driven case where a 3-item cart was off by one rupee; I checked the trace, reproduced it manually and logged it with the failing input. Result: it was fixed before release, and the team agreed to keep pricing tests in the pull request suite.
The classification tree method breaks the input domain into classifications (aspects like payment type, user type, device) and classes (values like card, UPI, wallet). These are drawn as a tree, and test cases are formed by picking one class from each classification in a combination table. It makes coverage visible, helps reviewers spot missing classes and combines well with pairwise selection to keep the number of test cases manageable.
19
What is the OWASP Top 10 and why should testers know it?
The OWASP Top 10 is a widely used list of the most critical web application security risks, published by an open community project. It includes broken access control, cryptographic failures, injection, insecure design, security misconfiguration, vulnerable components and authentication failures. Testers use it as a checklist to include basic security checks in their testing and to speak the same language as security teams.
20
Tell me about a time exploratory testing helped you find a critical bug.
Model answer: Situation: a new leave management module had passed all scripted cases. Task: I had a one-hour exploratory session. Action: I applied leave, then changed my manager in the profile while the request was pending. Result: the request disappeared for both managers and could never be approved. It was fixed before release, and the team added role-change scenarios to the regression suite.
Agree on scope: key user journeys and page templates, not every page. Day one, run automated scans on all templates to fix easy issues fast. Then do manual checks of each journey with keyboard, NVDA, VoiceOver on mobile, zoom and contrast. Log issues with WCAG references and severity, and get blockers fixed first. Retest fixes and deliver a conformance summary with known issues. Be honest if full compliance is not realistic.
24
What are the main testing differences between Android and iOS?
Android has many manufacturers, screen sizes and OS versions with custom skins, so device fragmentation and background-kill behaviour need more coverage. iOS has fewer devices but strict permission, privacy and store review rules. Navigation differs: Android has a system back button or gesture, iOS relies on in-app navigation and swipe back. Notification, permission and file-access behaviours also differ, so test cases need platform-specific steps.
25
A feature passed all your test cases, but users say it is confusing. What went wrong?
The tests verified the specification but did not validate the user experience. The test cases probably checked only functional outcomes and missed usability: unclear labels, too many steps or poor error messages. Next time, add usability heuristics to reviews, do exploratory sessions as a real user persona and involve users in UAT or demos. Log the current feedback as improvement items with clear examples.
A deep link opens a specific screen inside the app from a URL, email, SMS, QR code or notification. Test that each link opens the correct screen with the right data when the app is installed, closed or running in background. If the app is not installed, it should open the store page or a web fallback. Test links that need login, expired or invalid links, and that parameters cannot be tampered with to access other users' data.
27
What is JMeter and what does a basic test plan in it contain?
Apache JMeter is an open-source load testing tool, mainly for HTTP APIs and web pages. A basic test plan has a Thread Group that sets the number of virtual users, ramp-up time and loop count; samplers such as HTTP Requests; config elements like headers and CSV data; timers for think time; assertions on responses; and listeners or reports showing response times, throughput and errors. Run real load in non-GUI mode.
28
What is the difference between a test suite and a test run?
A transaction groups several database operations so they all succeed or all fail together, following ACID properties. To test rollback, force a failure in the middle of a multi-step action, for example a money transfer where the credit step fails after the debit. Then query both accounts and confirm neither balance changed and no partial records exist. Also check the user sees a clear error and can retry safely.
31
What is a performance bottleneck and where do bottlenecks usually occur?
A bottleneck is the component that limits the whole system's performance, so adding load beyond it increases response times or errors. Common places are slow database queries or missing indexes, too few database connections or threads, CPU or memory limits on servers, slow third-party APIs, large uncompressed files and chatty code that makes many small calls. Monitoring each layer during a test shows which resource saturates first.
Open the dialog with the keyboard. Focus should move into it, usually to the first field or heading, and the screen reader should announce it as a dialog with its title. Tab must stay inside the dialog while it is open and the page behind must not be reachable. Escape and the close button should close it, and focus must return to the button that opened it. Also check it reads correctly when zoomed.
33
What are ACID properties and how can a tester check them?
ACID means Atomicity, Consistency, Isolation and Durability. Atomicity: a transfer that debits one account must also credit the other, or neither happens; test by forcing a failure midway. Consistency: constraints still hold after the transaction. Isolation: two users updating together do not see each other's half-done work; test with two sessions. Durability: committed data survives a restart; test by committing, restarting the service and querying again.
A workload model describes the realistic load to simulate: which user journeys run, their mix in percentages, the number of concurrent users, request rates, think times, data volumes and the ramp-up pattern. For example, 60 percent browse, 25 percent search, 10 percent add to cart and 5 percent checkout at a peak of 2,000 concurrent users. It is built from production analytics and business forecasts, and wrong models produce meaningless results.
35
What sensitive data checks do you do in API responses and logs?
Check that responses return only needed fields, with no passwords, password hashes, full card numbers, full ID numbers or internal keys. Verify personal data is masked where required, for example only the last four digits of a phone number. Look at application and browser console logs for tokens or personal data. Check error responses do not expose stack traces, SQL or server versions. Raise any leak as a high-severity bug.
36
SQL: verify that soft-deleted users do not appear in the active users list.
-- the active list query used by the app should return 0 rows here
SELECT u.user_id, u.email
FROM users u
WHERE u.is_deleted = 1
AND u.user_id IN (
SELECT user_id FROM active_users_view
);
37
What is network condition testing for mobile apps?
It checks app behaviour on different networks: Wi-Fi, 4G, 5G, slow 2G or 3G, high latency, packet loss and no network. Also test switching between Wi-Fi and mobile data in the middle of an action, and airplane mode. The app should show loaders and clear offline messages, retry safely without duplicate actions and not freeze. Network throttling tools in developer options or proxy tools help simulate these conditions.
38
How do you capture logs from an Android device for a bug report?
Enable developer options and USB debugging, connect the phone and confirm it is visible with adb devices. Clear old logs with adb logcat -c, reproduce the bug, then save logs with adb logcat -d > crash_log.txt, filtering by the app package or error level if needed. For a crash, search for FATAL EXCEPTION. Attach the log, device model, OS version, app version and a screen recording to the bug.
39
What is localisation and internationalisation testing?
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.
Measuring speed and stability: load (expected users), stress (beyond limits), spike (sudden surge), endurance (long duration) and scalability testing.
Frequently asked questions
What do Software Testing interviews for ₹15–20 LPA roles test?
At ₹15–20 LPA the focus is usually on: deep expertise in a specialisation, architecture-level reasoning, metrics and business impact, leading small initiatives. This page's set is weighted towards Performance and Security Basics (7), Testing Types (5), Mobile Testing (5), Test Case Design (4).
Which companies pay ₹15–20 LPA for Software Testing roles?
Pay depends on company type, city, skills and negotiation, not only experience. Use the company-wise pages to practise for the specific employers you are targeting, and treat salary bands as a rough guide.
How should I prepare for a ₹15–20 LPA Software Testing interview?
Solve every question here without looking at the answer first, mark the ones you miss, save them to Last-Minute Revision and revise them the night before. Prepare two project stories that show impact at the level expected for ₹15–20 LPA.