Interview experience
Cognizant Interview Questions and Answers (2026)
Overview
Section titled “Overview”Cognizant hires freshers into three bands - GenC, GenC Pro, and GenC Elevate - through the same four-stage funnel, with the online assessment length, coding difficulty, and interview depth all scaling up by band.
Cognizant interview process at a glance
Section titled “Cognizant interview process at a glance”| Round | Duration | What they test |
|---|---|---|
| Communication Assessment | 20-30 min | AI-scored spoken/written English, listening, grammar |
| Aptitude Test | 30-45 min | Quantitative, logical reasoning, data interpretation, verbal |
| Technical Assessment | 45-60 min | Coding + fundamentals in your chosen skill cluster (scales by band) |
| Interview Panel (Tech + HR) | 30-45 min (GenC); up to ~2 hrs across two rounds for Elevate | Project discussion, CS fundamentals, fit |
Communication Assessment
Section titled “Communication Assessment”The first gate, and the one with the strictest cutoff - it’s scored by AI rather than a human interviewer, and a weak result ends the process before any technical round begins.
Common questions
- Listening-comprehension exercises with short audio clips
- Spoken-response prompts scored on fluency and grammar
- Grammar and sentence-correction MCQs
- Short extempore-style answers on a given topic
Aptitude Test
Section titled “Aptitude Test”A standard quant/reasoning/verbal screen. GenC-track papers run closer to 20 aptitude plus 15 verbal questions in about 25-30 minutes combined; higher bands see more questions and tighter time pressure.
Common questions
- Time and work, percentages, ratios (quantitative)
- Number series and coding-decoding (logical reasoning)
- Data interpretation from tables and graphs
- Reading comprehension and grammar/vocabulary (verbal)
Technical Assessment
Section titled “Technical Assessment”Coding and fundamentals in whichever skill cluster you registered for. Difficulty scales sharply by band: GenC gets 1-2 easy problems (sum of array elements, palindrome check), GenC Pro gets 2-3 medium problems (second-largest element, anagram check), and GenC Elevate gets 3 medium-hard problems (longest common subsequence, number of islands via BFS/DFS, coin change/DP).
Common questions
- Basic array/string problems - sum of array elements, palindrome check (GenC)
- Second-largest element in an array, anagram check (GenC Pro)
- Longest common subsequence, number of islands (BFS/DFS), coin change (DP) (GenC Elevate)
- Fundamentals questions in your chosen cluster (SQL syntax, Java OOPs, Python basics, or cloud concepts)
Round-by-round breakdowns for each band are on the Cognizant interview experience page.
Interview Panel (Technical + HR)
Section titled “Interview Panel (Technical + HR)”A single stage that folds a technical interview and an HR interview together for GenC and GenC Pro, but splits into two separate technical interviews plus a managerial/HR round for GenC Elevate.
Common questions
- OOPs pillars with examples; reverse a string; array vs linked list (GenC level)
- Implement binary search; explain sorting-algorithm complexity; DBMS normalization and SQL joins (GenC Pro level)
- Implement an LRU cache; find a loop in a linked list; design a simple URL shortener (GenC Elevate level)
- Process vs thread, deadlock, ACID properties, REST APIs (GenC Elevate CS-fundamentals round)
- Why Cognizant, willingness to relocate, and a time miscommunication caused a problem in a group project
Sample answer frameworks for the HR portion are on the Cognizant HR interview questions page.
GenC vs GenC Pro vs GenC Elevate: how the bands differ
Section titled “GenC vs GenC Pro vs GenC Elevate: how the bands differ”Cognizant doesn’t ask you to apply to a specific band - your OA performance and profile decide which track’s interview you get. GenC stays closest to programming fundamentals with a short OA and a single 30-minute interview. GenC Pro adds DBMS/OS depth and a longer coding section. GenC Elevate is the most competitive: a 120-minute OA with harder DSA, two separate technical interviews (one going into light system design), and a starting package that can run more than double GenC’s. Since the split happens off assessment performance rather than a separate application, the highest-leverage prep is pushing your coding-round score up rather than trying to pick a track.
Common technical interview questions and answers
Section titled “Common technical interview questions and answers”Q: How do you find the second largest element in an array?
Do it in a single pass with two variables, largest and second, both initialised to negative infinity. For each element: if it is bigger than largest, second becomes largest and largest becomes the element; else if it is bigger than second and not equal to largest, second becomes the element. That is O(n) time and O(1) space, and it beats the sort-then-take-index-n-minus-2 answer, which costs O(n log n). The two edge cases panels probe are arrays with fewer than two elements and arrays where the maximum repeats, such as 5, 5, 3 - the answer there is 3, not 5, if distinct values are required.
Q: How do you check whether two strings are anagrams?
Two strings are anagrams if they contain the same characters with the same counts. Reject immediately if the lengths differ, then build a frequency count of the first string and decrement it for each character of the second; if any count goes negative or a character is missing, they are not anagrams. This is O(n) time and O(k) space for k distinct characters. Sorting both strings and comparing also works and is easier to write, but it costs O(n log n) - mention both and say which you would ship.
Q: How do you detect a loop in a linked list?
Use Floyd’s cycle-detection algorithm, also called the tortoise and hare. Move a slow pointer one node at a time and a fast pointer two nodes at a time; if the list ends, there is no cycle, and if the two pointers ever meet, there is one. It uses O(n) time and O(1) space. To find where the cycle starts, reset one pointer to the head after the meeting and advance both one step at a time - they meet at the cycle entry. The hash-set alternative is simpler but uses O(n) memory, which is the trade-off the interviewer wants you to name.
Q: How does the coin change problem work with dynamic programming?
To find the fewest coins summing to an amount, build an array dp where dp[i] is the minimum coins needed for amount i. Initialise dp[0] = 0 and everything else to infinity, then for every coin c and every amount i from c up to the target, set dp[i] = min(dp[i], dp[i - c] + 1). The answer is dp[amount], or -1 if it is still infinity. Complexity is O(amount times number of coins) time and O(amount) space. Greedy - always taking the largest coin - is wrong for coin sets like 1, 3, 4, where the amount 6 needs two 3s, not 4 plus 1 plus 1.
Q: How do you solve the number of islands problem?
Treat the grid as a graph where each land cell connects to its four neighbours. Scan every cell; when you hit unvisited land, increment your island counter and run a BFS or DFS from that cell, marking every reachable land cell as visited (either in a separate visited array or by overwriting the cell with water). Each cell is visited once, so it is O(rows times columns) time. DFS uses recursion stack space up to the size of the largest island, so for very large grids an explicit stack or BFS queue avoids stack overflow - a follow-up Elevate panels do ask.
Q: What is normalization in DBMS, and what are 1NF, 2NF and 3NF?
Normalization organises tables to remove redundancy and update anomalies. First normal form requires atomic values - no comma-separated lists or repeating groups in a column. Second normal form applies when the primary key is composite and requires that every non-key column depend on the whole key, not just part of it. Third normal form removes transitive dependencies: a non-key column must not depend on another non-key column, so storing dept_name alongside dept_id in an employee table violates 3NF and belongs in a separate departments table. The trade-off is that heavy normalization means more joins, so reporting systems often denormalize deliberately.
Q: What is the difference between a process and a thread, and what causes deadlock?
A process is an independent program in execution with its own memory space; a thread is a unit of execution inside a process that shares that process’s heap and open files while keeping its own stack and registers. So threads are cheaper to create and switch between, but a bug in one thread can corrupt shared state, whereas processes are isolated. Deadlock happens when four conditions hold at once: mutual exclusion, hold and wait, no preemption, and circular wait. Break any one of them to prevent it - the usual practical fix is imposing a global lock-ordering rule so a circular wait cannot form.
Q: What makes an API RESTful, and which HTTP methods are idempotent?
A REST API models resources as URLs and uses HTTP methods for actions on them, is stateless (each request carries everything needed, with no server-side session), and returns standard status codes - 200 for success, 201 for created, 400 for a bad request, 401 for unauthenticated, 404 for not found, 500 for a server error. GET, PUT and DELETE are idempotent: repeating the same call leaves the server in the same state. POST is not, which is why submitting a payment form twice can create two payments. GET is additionally safe because it must not modify anything.
Frequently asked questions about Cognizant interviews
Section titled “Frequently asked questions about Cognizant interviews”What is Cognizant’s GenC interview process for freshers?
Cognizant’s GenC hiring runs 4 stages: 1. Communication Assessment - an AI-scored test of spoken/written English, listening, and grammar; this has the strictest cutoff and eliminates the most candidates. 2. Aptitude Test - quantitative reasoning, logical reasoning, data interpretation, and verbal ability. 3. Technical Assessment - coding and fundamentals in a ‘skill cluster’ you pick at registration (e.g., Java, SQL, HTML/CSS/JS, Python, or Cloud). 4. Interview Panel - a technical interview (project discussion, CS fundamentals, sometimes light coding) followed by an HR interview.
What is the difference between GenC, GenC Pro, and GenC Elevate?
All three run through the same four-stage funnel, but the online assessment and interview scale up with the band. GenC: a 60-minute OA (20 aptitude + 15 verbal questions + 1-2 easy coding problems), a 30-minute technical interview on programming basics and project discussion, and a 20-minute HR round - packages cluster around Rs 4 LPA. GenC Pro: a 90-minute OA with 2-3 medium coding problems, a 45-minute technical interview covering DSA fundamentals plus DBMS/OS, and a 25-minute HR round - packages run roughly Rs 6-6.5 LPA. GenC Elevate: a 120-minute OA with 3 medium-hard coding problems (DP, graph traversal), two technical interviews (advanced DSA plus system-design basics, then CS fundamentals and a project deep-dive), and a 30-minute HR/managerial round - packages run roughly Rs 6.5-9 LPA.
Why does the Communication Assessment carry so much weight at Cognizant?
It’s the first gate and is AI-scored, so it filters out a large share of candidates before any technical evaluation happens. Because Cognizant staffs global client accounts where consultants communicate directly with clients, clear spoken and written English is treated as a baseline hiring requirement, not a soft extra.
What is a ‘skill cluster’ in Cognizant’s technical assessment?
At registration, candidates choose a preferred technology track - options typically include Java, ANSI SQL, HTML/CSS/JavaScript, Python, or Cloud fundamentals - and the Technical Assessment round tests that specific track rather than a generic CS syllabus. Your cluster choice can also influence which project/team you’re eventually staffed on.
How many rounds are there in the Cognizant interview?
Four stages for every band: Communication Assessment, Aptitude Test, Technical Assessment (skill cluster), and an Interview Panel that folds a technical interview and an HR interview into the same stage. GenC Elevate candidates typically face two separate technical interviews within that panel stage rather than one, given the deeper DSA and system-design expectations.
How should I prepare for Cognizant interviews?
Practice spoken English and grammar for the Communication Assessment, revise quantitative and logical reasoning for the Aptitude round, pick a skill cluster you’re genuinely comfortable in and go deep on it, and prepare to discuss any academic or internship projects clearly for the interview panel. There’s no negative marking in the assessments, so attempt every question. If you’re aiming for GenC Pro or Elevate, push your DSA practice well beyond array/string basics into sorting, DP, and graph traversal.