Cognizant Interview Questions
Cognizant hires freshers under the GenC banner, but GenC, GenC Next and GenC Elevate are not simply score bands of one test — they involve different assessments and different interview depth, which is the part candidates most often misunderstand. This page covers the tiers and what each actually requires, the assessment pattern, and the technical and HR questions that recur.
77 questions with concise, interview-ready answers.
Process and the GenC tiers
1. What is the difference between GenC, GenC Next and GenC Elevate? Fresher
GenC is the standard fresher programme. GenC Next is the premium engineering track — it involves an advanced coding assessment and a harder technical interview, and pays substantially more. GenC Elevate sits between them for candidates who perform strongly but not at GenC Next level. The tiers differ in which assessment you sit, not only in the score you achieve, so check which one your drive is for.
2. What are the rounds in the Cognizant hiring process? Fresher
An online assessment covering aptitude, logical reasoning, verbal ability and a programming or coding section, followed by a technical interview and an HR interview. For GenC Next there is an additional advanced coding round and the technical interview goes materially deeper into data structures and problem solving.
3. What is the eligibility criteria for Cognizant? Fresher
Typically a full-time recognised degree, a minimum aggregate across academics, no active backlogs at the time of joining, and limits on education gaps. GenC Next applies stricter filters than standard GenC. As always the notification for your drive governs, since these have changed between cycles.
4. How should you prepare differently for GenC Next? Fresher
Treat it as a product-company-style interview rather than a services one. That means real data-structures and algorithms practice — arrays, strings, hash maps, trees, recursion and basic dynamic programming — with attention to complexity analysis, not just working code. Standard GenC preparation weighted toward aptitude and definitions is not sufficient for this track.
5. What is GenC Pro and how does it differ from the other tracks? Fresher
GenC Pro sits at the top of the fresher family, aimed at candidates with demonstrable engineering depth, and carries the highest package and the hardest selection bar of the group. In practice the tracks form a ladder — GenC, GenC Elevate, GenC Next, GenC Pro — where each step adds assessment difficulty and interview depth rather than simply raising a cut-off on the same test. Track names and criteria have been revised between hiring cycles, so confirm what your specific drive is actually offering.
6. What does the Cognizant online assessment cover, section by section? Fresher
Typically quantitative aptitude, logical reasoning and verbal ability, plus a programming section that mixes output-prediction and concept questions with one or more coding problems. Sections are individually timed, so pace is enforced within each one and cannot be traded across them. On the higher tracks the coding component carries more weight and the problems move from single-loop string work toward genuine data-structure problems judged on hidden test cases.
7. What is the GenC Next advanced coding round like? Fresher
Expect two or three problems in a fixed window, scored against hidden test cases rather than the sample output, at roughly easy-to-medium competitive-programming difficulty: arrays and strings, hash maps, sorting, sometimes a tree or a simple dynamic-programming state. Partial credit for passing some tests is common, so writing a correct brute force before attempting the optimal solution is a sound strategy. Your approach and its complexity get discussed afterwards, so know the running time of what you submitted.
8. What changes in the technical interview between GenC and GenC Next? Fresher
At GenC level the interview is largely definitional — OOP concepts, DBMS terms, your project, one small program on paper or a shared editor. At GenC Next you are asked to write code and then reason about it: the complexity of your approach, why you chose one data structure over another, which edge cases break it, and how it behaves as input grows. The project discussion also shifts from what you built to why you built it that way.
9. Can you apply to Cognizant again after being rejected? Fresher
Yes, after a cooling-off period, which has commonly run to a few months and depends on the stage you reached and whether the drive was on-campus or off-campus. The recruiter or the drive notification is the authority, since this has changed between cycles. Use the gap to close the specific gap that failed you — repeating identical assessment practice and expecting a different score rarely works.
Data structures and algorithms
10. What is a linked list and what are its variants? Fresher
A sequence of nodes where each holds data and a pointer to the next. A singly linked list points forward only; a doubly linked list also points back, allowing O(1) removal given a node; a circular list has its last node point to the first. Insertion and deletion at a known position are O(1), but access and search are O(n).
11. How do you detect a cycle in a linked list? Fresher
Use Floyd's cycle detection: advance a slow pointer one node and a fast pointer two nodes per step; if they ever meet there is a cycle, and if the fast pointer reaches null there is not. That is O(n) time and O(1) space. A hash set of visited nodes also works but costs O(n) space, so mention both and say why you prefer the pointers.
12. What is a binary search tree and what is its worst case? Fresher
A binary tree where every left descendant is smaller than the node and every right descendant is larger, giving O(log n) search, insert and delete when balanced. The worst case is O(n): inserting sorted data degenerates the tree into a linked list. That is precisely why self-balancing variants such as AVL and red-black trees exist.
13. What is the difference between BFS and DFS? Fresher
Breadth-first search explores level by level using a queue, so it finds the shortest path in an unweighted graph, at the cost of holding a whole frontier in memory. Depth-first search follows one branch to its end using a stack or recursion, using less memory on wide graphs and suiting cycle detection and topological sorting. Both are O(V + E).
14. What is a hash table and how are collisions handled? Fresher
A hash table maps a key to a bucket index via a hash function, giving average O(1) lookup. Collisions — two keys hashing to the same bucket — are handled either by chaining, where each bucket holds a list or tree of entries, or by open addressing, where the entry is placed in another probed slot. Worst case degrades to O(n) with a poor hash function.
15. What is dynamic programming? Fresher
A technique for problems with overlapping subproblems and an optimal substructure: solve each subproblem once and reuse the result, either top-down with memoised recursion or bottom-up with a table. It converts exponential recursive solutions into polynomial ones — the Fibonacci sequence, knapsack and longest common subsequence are the standard examples.
16. How would you find the first non-repeating character in a string? 2–5 yrs
Two passes over the string: the first counts occurrences in a hash map or fixed-size array, the second returns the first character whose count is one. That is O(n) time and O(k) space. A single-pass variant storing first-seen indices also works but is more code for no asymptotic gain — say so rather than reaching for it reflexively.
17. How would you check whether a string is a palindrome? Fresher
Use two pointers, one at each end, comparing characters and moving inward until they cross — O(n) time and O(1) extra space. Reversing the string and comparing also works but allocates a second string for no benefit. Clarify the requirements before you write anything: whether case matters and whether punctuation and spaces should be skipped, because the interviewer is often waiting to see whether you ask.
18. Given an array and a target, how do you find two numbers that sum to the target? Fresher
One pass with a hash map: for each element check whether the target minus that element has already been seen, and if not, store the element with its index. That is O(n) time and O(n) space. The nested-loop version is O(n squared). If the array is already sorted, two pointers converging from both ends give O(n) time in O(1) space, which is the better answer when sorting is free.
19. How do you find the missing number in an array containing 1 to n? Fresher
Compute the expected sum with n times n plus one over two, subtract the actual array sum, and the difference is the missing value — O(n) time and O(1) space. XOR of all indices with all values gives the same answer and avoids any overflow concern on very large n, which is the follow-up interviewers like. Sorting the array first works but costs O(n log n) for no gain.
20. How do you check whether two strings are anagrams? Fresher
Count character frequencies in one pass over each string, using a fixed-size array for a known alphabet or a hash map otherwise, then compare the counts — O(n) time. Sorting both strings and comparing is easier to write but O(n log n). State your assumptions first: case sensitivity, whitespace handling, and whether the input is restricted to ASCII, since Unicode input breaks the fixed-array version.
21. How do you reverse a linked list? Fresher
Iteratively with three pointers — previous, current and next. Save the current node's next, point current at previous, then advance previous to current and current to the saved node, until current is null; previous is the new head. That is O(n) time and O(1) space. The recursive version is shorter but consumes O(n) stack, which matters on long lists, so mention both and say which you would ship.
22. How do you find the middle element of a linked list in a single pass? Fresher
Advance a slow pointer one node and a fast pointer two nodes per iteration; when the fast pointer reaches the end, the slow pointer is at the middle. O(n) time, O(1) space, one traversal. Decide up front which of the two middles you return for an even-length list, because that off-by-one is the usual reason a correct-looking solution fails a hidden test.
23. What are the binary tree traversals and when is each useful? Fresher
Inorder visits left, node, right, and on a binary search tree emits values in sorted order. Preorder visits the node first, which suits copying or serialising a tree. Postorder visits both children before the node, which suits deleting a tree or evaluating an expression bottom-up. Level order uses a queue and gives you the tree row by row. All four are O(n) in time.
24. How do you find the height of a binary tree and check whether it is balanced? 2–5 yrs
Height is one plus the greater of the two subtree heights, with an empty subtree counted as zero or minus one depending on your convention — a straightforward O(n) recursion. Checking balance naively recomputes the height at every node and costs O(n squared); returning height and a balance flag together in one bottom-up pass brings it back to O(n). That optimisation is usually the entire point of the question.
25. When does recursion cause a stack overflow, and what can you do about it? Fresher
Every call holds a frame until it returns, so recursion depth proportional to input size will exhaust the stack — a recursive walk over a million-node list or a degenerate tree is the standard example. The fixes are converting to an iterative loop with an explicit stack, or restructuring into tail recursion where the compiler eliminates the frame. Note that Python does not eliminate tail calls and enforces its own recursion limit.
26. Why is naive recursive Fibonacci slow, and how does memoisation fix it? Fresher
The naive version recomputes the same subproblems exponentially often, giving roughly two to the power of n calls. Memoising results in a map or array makes each of the n subproblems compute once, dropping the running time to O(n) with O(n) space. The bottom-up version keeps only the previous two values and runs in O(n) time with O(1) space, which is the answer to reach for when asked to optimise further.
27. How do you find the maximum sum of a contiguous subarray? 2–5 yrs
Kadane's algorithm: walk the array tracking the best sum ending at the current position, which is either the current element alone or the current element added to the previous best, and keep the maximum seen overall. O(n) time and O(1) space. The trap is an array of all negative numbers — initialising the answer to zero wrongly returns zero, so initialise from the first element instead.
28. How do you check whether a string of brackets is balanced? Fresher
Push each opening bracket onto a stack; on a closing bracket, pop and confirm the popped bracket is the matching type, failing immediately if the stack is empty. At the end the string is balanced only if the stack is also empty — omitting that final check is the most common bug in this problem. O(n) time and O(n) space in the worst case.
Programming and OOP
29. What is inheritance and what problem does it cause? Fresher
Inheritance lets a class derive fields and behaviour from a parent, enabling reuse and polymorphism. The problem is tight coupling: a subclass depends on its parent's internals, so a parent change can break it, and deep hierarchies become hard to reason about. Composition is often the better default, which is worth saying — it shows you know the trade-off, not just the definition.
30. What is polymorphism, with an example? Fresher
One interface serving multiple underlying types. Compile-time polymorphism is method overloading; runtime polymorphism is overriding, where the actual object's method is chosen at execution. The concrete example: a Shape reference calling area() runs Circle's or Rectangle's implementation depending on the real object, so calling code needs no knowledge of the subtype.
31. What is encapsulation and how is it enforced? Fresher
Bundling data with the methods that operate on it and restricting direct external access to the state, so invariants can be maintained in one place. It is enforced with access modifiers — private fields exposed through methods that validate changes. The point is not hiding for its own sake but ensuring an object cannot be put into an invalid state from outside.
32. What is the difference between static and instance members? Fresher
A static member belongs to the class and exists once regardless of how many objects are created; an instance member exists per object. Static methods cannot access instance state directly because there is no particular object involved. Static mutable state is a common source of bugs in concurrent code, which is a good caveat to add.
33. What is garbage collection? Fresher
Automatic reclamation of memory no longer reachable from live references, removing the need for manual frees and preventing most leaks and dangling pointers. The cost is that collection happens at times you do not control, which can introduce pauses. You can make objects eligible for collection but cannot force a collection to happen at a chosen moment.
34. What is the difference between abstraction and encapsulation? Fresher
Abstraction is the design decision about what to expose — presenting a simplified model of a thing and omitting detail the caller does not need. Encapsulation is the mechanism that protects the internals once that decision is made, keeping state private and mediating access through methods. Abstraction is what the caller sees; encapsulation is how you enforce it. Interviewers ask this precisely because the two are so often blurred into one answer.
35. What is the difference between an abstract class and an interface? Fresher
An abstract class can hold fields, constructors and concrete methods, and a class can extend only one, so it fits a family of closely related types sharing implementation. An interface declares a contract that entirely unrelated types can implement, and a class can implement many. Default methods on modern Java interfaces narrow the gap, but only an abstract class can hold instance state — that is the distinction that survives.
36. What is the difference between method overloading and method overriding? Fresher
Overloading means several methods in one class share a name but differ in parameters, resolved by the compiler from the argument types. Overriding means a subclass replaces an inherited method with an identical signature, resolved at runtime from the actual object type. Changing only the return type overloads nothing, and marking a method final, static or private prevents overriding — both are standard follow-up questions.
37. What is the difference between checked and unchecked exceptions? Fresher
Checked exceptions must be declared or handled and describe conditions a caller could reasonably recover from, such as a missing file or a failed network call. Unchecked exceptions extend RuntimeException and usually signal programming errors — a null dereference, a bad index — that should be fixed rather than caught. The mistake to avoid is catching an exception and leaving the block empty, which hides the failure and makes the eventual bug far harder to locate.
38. Why should you never compare strings with == in Java? Fresher
Because == compares object references rather than contents. Two strings with identical text can be different objects: compile-time literals are interned into a shared pool, but strings built at runtime by concatenation or read from input are not. Use equals(), or equalsIgnoreCase() when case is irrelevant. The reason this bug survives testing is that it appears to work for the literal cases developers usually test with.
39. What is a race condition and how do you prevent one? 2–5 yrs
A race condition is when the outcome depends on the unpredictable interleaving of threads touching shared mutable state — the classic case being read-modify-write on a counter, where two increments produce one. Prevent it by removing the sharing, making the state immutable, or serialising access with a lock, synchronised block or atomic type. Locks bring their own risk: acquire multiple locks in a consistent global order or you have traded a race for a deadlock.
40. What is the difference between a shallow copy and a deep copy? 2–5 yrs
A shallow copy duplicates the outer object but keeps references to the same nested objects, so mutating a nested object through one copy is visible through the other. A deep copy recursively duplicates the nested objects, leaving the two fully independent. Deep copies cost time and memory and have to cope with reference cycles, which is why most default clone implementations are shallow — and why that catches people out.
DBMS and SQL
41. What is the difference between a clustered and a non-clustered index? Fresher
A clustered index determines the physical order of rows in the table, so there can only be one, and it makes range scans on that column fast. A non-clustered index is a separate structure holding the key and a pointer back to the row, so you can have several, but a lookup may need an extra step to fetch the row.
42. What is a foreign key? Fresher
A column or set of columns referencing the primary key of another table, enforcing referential integrity — you cannot insert a child row pointing at a parent that does not exist, and deleting a referenced parent is blocked or cascaded depending on the rule defined. It is how a relational database guarantees relationships stay valid.
43. What is the difference between UNION and UNION ALL? Fresher
UNION combines result sets and removes duplicate rows, which requires a sort or hash and therefore costs time. UNION ALL concatenates without deduplicating and is faster. If you know the sets are disjoint, UNION ALL is the correct choice — using UNION by default is a common and avoidable performance mistake.
44. What is a stored procedure and when is it useful? 2–5 yrs
A named, precompiled block of SQL stored in the database and invoked by name, optionally with parameters. It reduces round trips, centralises logic close to the data, and can be granted permissions independently. The trade-off is that business logic in the database is harder to version-control and test than logic in the application.
45. What are the types of joins in SQL? Fresher
An inner join returns only rows that match on both sides. A left join returns every row from the left table with NULLs where the right has no match, and a right join is its mirror. A full outer join returns unmatched rows from both sides. A cross join returns the Cartesian product. A self join joins a table to itself, which is how a hierarchy such as employee and manager in one table is queried.
46. What is normalization, and when would you deliberately denormalise? Fresher
Normalization organises tables so each fact lives in exactly one place: first normal form requires atomic values, second removes partial dependencies on part of a composite key, third removes transitive dependencies between non-key columns. Denormalization reintroduces redundancy — a duplicated column or a precomputed total — to avoid joins on a read-heavy path. That is a deliberate trade of write complexity for read speed, and it belongs in reporting and analytics tables rather than the transactional core.
47. What are the ACID properties? Fresher
Atomicity: the transaction is all or nothing. Consistency: it moves the database between valid states with constraints respected. Isolation: concurrent transactions do not observe each other's intermediate state, to the extent the isolation level guarantees. Durability: once committed, the change survives a crash, usually because it was written to a log before being acknowledged. Isolation draws the follow-up questions, so know that read-committed and repeatable-read differ in which anomalies they permit.
48. What is the difference between WHERE and HAVING? Fresher
WHERE filters rows before aggregation and cannot use aggregate functions; HAVING filters the groups produced by GROUP BY and is the only place an aggregate condition can appear. Because filtering earlier means grouping fewer rows, any condition not involving an aggregate belongs in WHERE for performance. A single query can legitimately use both clauses together.
49. What is the difference between DELETE, TRUNCATE and DROP? Fresher
DELETE removes selected rows, logs each one, fires triggers and can be rolled back. TRUNCATE removes every row in one bulk operation, is far faster, generally resets identity values, and cannot be filtered with a WHERE clause. DROP removes the table structure altogether. The usual follow-up is which are DML and which are DDL: DELETE is DML, TRUNCATE and DROP are DDL, which is why their rollback behaviour differs by database.
50. Write a query to find the second highest salary. 2–5 yrs
Select the maximum salary from the table where salary is less than the overall maximum — a simple nested subquery that handles ties at the top correctly and returns NULL when no second value exists. The window-function alternative applies DENSE_RANK ordered by salary descending and filters for rank two, which extends naturally to the Nth highest and to a per-department answer with PARTITION BY. Avoid offering ORDER BY with OFFSET as your only answer, since it breaks when the top salary is duplicated.
51. What is an index, and why not index every column? 2–5 yrs
An index is a separate sorted structure, usually a B-tree, that lets the database locate rows without scanning the table, turning many lookups from O(n) into O(log n). Every index has to be maintained on insert, update and delete and consumes storage, so over-indexing slows writes and wastes space. Indexes also go unused when a query wraps the column in a function, or when the column is so unselective that a table scan is cheaper.
52. What is the difference between a primary key and a unique key? Fresher
Both enforce uniqueness. A primary key additionally defines row identity, cannot contain NULL, and there is exactly one per table — it is what foreign keys reference. A unique key enforces a business rule such as one account per email address, permits a NULL in most databases, and a table can carry several. Choosing a natural business column as the primary key is risky because business values change; surrogate keys avoid that problem.
Operating systems and networking
53. What is a process and what states does it move through? Fresher
A process is a program in execution with its own address space, open file handles and register state. It moves between new, ready, running, waiting and terminated: the scheduler promotes it from ready to running, an I/O request pushes it to waiting, and completion of that I/O returns it to ready. Every switch between processes costs a context switch that saves and restores state, which is pure overhead — so excessive switching hurts throughput.
54. What are the common CPU scheduling algorithms? Fresher
First-come first-served is trivial but suffers convoy effects when one long job blocks short ones. Shortest job first minimises average waiting time but requires burst times you usually do not have and can starve long jobs. Round robin gives each process a fixed time slice, bounding response time for interactive systems, with the slice size trading responsiveness against switch overhead. Priority scheduling risks starvation unless waiting processes are aged upward.
55. What is the difference between a semaphore and a mutex? 2–5 yrs
A mutex is a lock owned by the thread that acquired it, and only that thread may release it, enforcing mutual exclusion over one resource. A semaphore is a signalling counter permitting up to N concurrent holders, and any thread may signal it, which suits limiting a pool or coordinating a producer and consumer. A binary semaphore resembles a mutex but has no ownership rule, and that difference matters for correctness and for priority inheritance.
56. What is thrashing, and what do page replacement algorithms do? Fresher
Thrashing is when the combined working set of running processes exceeds physical memory, so the system spends most of its time servicing page faults instead of executing instructions and throughput collapses while the machine looks busy. Page replacement algorithms choose which page to evict on a fault: FIFO is cheap but can evict a hot page, LRU approximates the ideal well, and the theoretically optimal policy needs future knowledge and exists only as a benchmark.
57. When would you choose UDP over TCP? Fresher
When timeliness matters more than completeness and the application can tolerate or repair loss itself: live voice and video, online gaming, telemetry, and DNS queries small enough to fit one datagram. TCP retransmission and in-order delivery mean one lost packet stalls everything queued behind it, which is worse than a single dropped video frame. UDP gives you no ordering, acknowledgement or congestion control, so anything you need from that list you have to build yourself.
58. How does DNS resolution work? Fresher
The resolver checks local caches first — the browser, then the operating system, then the configured recursive resolver. On a miss, the recursive resolver asks a root server for the top-level domain nameservers, those for the domain's authoritative nameserver, and that server for the record itself. The answer is cached at each level for the record time-to-live, which is why a DNS change takes time to propagate and why lowering the TTL before a migration is standard practice.
59. What do the common HTTP status codes mean, and what does HTTPS add? Fresher
Two hundred is success; three hundred and one and three hundred and two are permanent and temporary redirects; four hundred and one means unauthenticated; four hundred and three means authenticated but not permitted; four hundred and four means not found; five hundred is a server-side failure. The rule of thumb is that four-hundreds blame the request and five-hundreds blame the server. HTTPS wraps the same protocol in TLS, adding encryption in transit, integrity, and authentication of the server through its certificate.
Aptitude and verbal ability
60. What does the Cognizant aptitude section cover and how tight is the timing? Fresher
Quantitative topics centre on percentages, ratio and proportion, averages, time and work, time speed and distance, simple and compound interest, and basic probability and permutations. Reasoning covers series, coding-decoding, seating arrangements, blood relations and syllogisms. Sections are separately timed, so roughly a minute per question is the working assumption. The binding constraint is pace, and practice exists to make accuracy survive that pace.
61. Two pipes fill a tank in 20 and 30 minutes and a third empties it in 60. How long to fill? Fresher
Convert everything to a per-minute rate and treat the outlet as negative work. The rates are 1/20 plus 1/30 minus 1/60, which over a common denominator of 60 is 3 plus 2 minus 1, giving 4/60 or 1/15 of the tank per minute. The tank therefore fills in 15 minutes. Every pipes-and-cisterns question is this same reciprocal addition with outflow subtracted.
62. Two dice are thrown. What is the probability that the sum is 8? Fresher
The sample space is 36 equally likely outcomes. The pairs summing to eight are 2 and 6, 3 and 5, 4 and 4, 5 and 3, and 6 and 2 — five outcomes, so the probability is 5/36. The mistake to avoid is treating 2 and 6 as the same outcome as 6 and 2: the dice are distinguishable, and collapsing them corrupts both the count and the denominator.
63. The ratio of two ages is 4 to 3 and in six years it will be 6 to 5. What are the ages? Fresher
Write the current ages as 4x and 3x rather than as two independent unknowns — that single step is what makes ratio problems fast. Then 4x plus 6 over 3x plus 6 equals 6 over 5, so 20x plus 30 equals 18x plus 36 and x is 3, giving ages of 12 and 9. Always add the same number of years to both people; adding to only one is the standard slip.
64. How do you approach coding-decoding and number series questions? Fresher
For coding-decoding, map letters to positions one to twenty-six and look for a constant shift, a reversal, or a shift that varies with position — checking the first and last letters usually reveals which. For number series, test first differences, then ratios, then second differences, then squares and cubes with an offset, and check whether alternate terms form two interleaved series. Cap your time: these resolve inside a minute or should be flagged and left.
65. How is the verbal ability section structured and how should you prepare? Fresher
Expect reading comprehension, sentence correction and error spotting, para jumbles, sentence completion and vocabulary in context. The highest-return preparation is the small set of grammar rules that recur — subject-verb agreement, tense consistency, misplaced modifiers, articles and prepositions — because they generate most of the error-spotting marks. For comprehension, read the questions first and scan for the relevant paragraph rather than reading the passage end to end.
HR and behavioural
66. Tell me about yourself. Fresher
Sixty to ninety seconds covering your background, one or two things you actually built, the resulting skills, and why this role. Lead with your strongest concrete item. This is the only fully predictable question in the round, so it should be the most polished answer you give.
67. Why Cognizant? Fresher
Name something specific — the GenC track you are targeting and why it suits your strengths, the healthcare and financial-services domain depth the company is known for, or the structured entry programme. An answer that shows you know which track you are interviewing for already distinguishes you from most candidates.
68. Tell me about a time you worked in a team. Fresher
Use the STAR structure — situation, task, action, result — and keep the action on what you personally did. Pick an example with a genuine complication, such as a teammate dropping out or a disagreement about approach, because a story where nothing went wrong demonstrates nothing about how you work.
69. How do you keep your technical skills current? Fresher
Be specific and verifiable: what you are learning now, where, and what you built with it. "I follow tech news" is not an answer. One small project finished recently is the strongest possible evidence here, and it gives the interviewer something concrete to ask about.
70. Are you willing to relocate and work in any technology? Fresher
Answer honestly. In a services business, allocation determines both your location and your stack, frequently after training, so this is a real question rather than a formality. State any genuine constraint now alongside the flexibility you do have.
71. Why are you leaving your current job? 2–5 yrs
Frame it toward what you are moving to — scope, technology, domain — rather than against your current employer. Criticising a present team or manager is the single most damaging answer in an HR round because it predicts how you would talk about this one. Keep it short and forward-looking.
72. What are your strengths and weaknesses? Fresher
Choose one strength relevant to the job and prove it with a specific instance instead of an adjective. For the weakness, pick something genuine, bounded and improving, and describe the concrete habit you use to manage it. The rehearsed answer about caring too much is transparent, and interviewers read it as evidence you would not admit a problem on a project either.
73. Tell me about a time you failed or made a mistake. Fresher
Pick a real failure with a visible consequence, state your own part in it without distributing blame, then spend most of the answer on what you changed afterwards and what happened the next time. The interviewer is checking whether you can own an error, because someone who cannot will hide the next one. A trivial example damages the answer as much as having no example at all.
74. How do you handle pressure and tight deadlines? Fresher
Answer with your actual method: list the work, identify what is genuinely on the critical path, cut or defer the rest, and flag the risk early to whoever depends on it. Then give one example of doing exactly that and what the outcome was. Claiming you work well under pressure without naming a mechanism sounds like tolerance for chaos rather than management of it.
75. What if you are allocated to testing or support instead of development? Fresher
Say honestly that you would take it and do it well, and mean it — allocation follows client demand and most people move between roles across their first few years. It helps to show you know these roles are technical: automation testing, production support and platform work all build transferable skill. A candidate who treats non-development allocation as a demotion is a retention risk, and this question exists to find that out.
76. What do you know about Cognizant? Fresher
Know the shape of the business rather than a slogan: a large IT services and consulting firm with particularly deep presence in healthcare and financial services, headquartered in the United States with the bulk of its delivery workforce in India, organised into industry-aligned practices. Then connect one of those facts to why you are applying. Reciting the founding year and revenue with no link to your own interest adds nothing.
77. Do you have any questions for us? Fresher
Ask something only this interviewer can answer: what the first six months look like on a typical project, how allocation and training work after joining, what distinguishes the people who do well on their team, or what technology their current project runs on. Having no questions reads as low interest. Leave compensation and leave policy for the offer conversation unless HR raises it first.
Get these answered live in your real interview
NostrobeAI is a real-time AI interview copilot — it hears the question and drafts a strong answer on your screen, invisible on Zoom, Meet, and Teams. One-time pricing, no subscription.
Try NostrobeAI free