Capgemini Interview Questions and Answers

Answer

As of 2026, candidates typically describe the fresher screen as several distinct blocks rather than one paper: a behavioural or game based component in the cycles where it is used, quantitative and logical reasoning, a pseudocode section, technical multiple choice questions, and a separate written English test. The exact composition and order have varied across cycles and across campus and off campus drives, so read your instruction screen rather than trusting a coaching PDF. The filter effect comes from that spread.

A candidate who spends six weeks on percentages and seating arrangements is genuinely well prepared for one block and completely unprepared for three others. Pseudocode in particular catches people out because it asks you to read unfamiliar syntax and predict output rather than write a program, which is a different skill from solving problems in your own language. The English test catches out candidates who have never typed a structured essay under a timer.

What the panel probes later, when they ask you about the test in the interview, is whether you understood the structure or simply showed up. A weak answer is 'it had aptitude and some coding'. A strong answer names the blocks you attempted, says where you were strong and weak, and does not inflate a score that is already on their system.

Answer

Gamified and behavioural screens measure traits and working style rather than knowledge, and they are usually built around consistency. You cannot revise for them, but you can avoid the three ways candidates damage themselves. The first is trying to guess the employer's ideal profile and answering as that person, which fails because the same trait is probed several times in different wordings and an invented profile produces contradictory responses that the scoring picks up.

The second is overthinking timed reaction or pattern tasks, where the instrument is measuring how you behave under mild pressure, so hesitating to look clever makes your profile noisier, not better. The third is treating the section as unimportant and clicking through it, which produces a flat, uninformative profile. The practical approach is to answer quickly and honestly, keep the same self across the whole section, and read the instructions properly before the first item because these formats are unfamiliar and most of the confusion is procedural rather than psychological.

If you are asked about it in the interview, describe it factually and neutrally. What a panel is probing there is self awareness. A weak answer is 'it was just some games, I do not think it meant anything'. A strong answer says what you think it was measuring and where you would place yourself honestly.

Answer

Assume that sections are individually timed and that once a section closes you cannot go back into it, because that is how these assessments have typically been run and it dictates your entire strategy. Budget your time per question before the section opens and enforce it mechanically. Set a hard walk away rule: if a question has taken more than about ninety seconds and you cannot see the path to the answer, mark it and move on.

The most common way a capable candidate fails is not weakness, it is spending seven minutes on one number system question and then rushing eight easy ones in the last four minutes. Work in two passes. Sweep the section first, taking everything you can solve confidently, then come back with the remaining time for the heavy items.

If a section closes with questions unattempted, do not carry it into the next section emotionally, because the next block is scored independently and a demoralised candidate loses far more marks in the following twenty minutes than they lost in the last two. Practise on a timer for at least ten full sections before test day. Pacing is a trained reflex, not a decision you can make live while a clock is running.

Answer

Capgemini has in several cycles run a separate higher band assessment marketed under the Exceller banner, sitting above the standard Analyst entry route. The essential difference is that it is a harder, separately scheduled test aimed at candidates who want the higher package, and clearing it routes you into a different offer rather than simply getting you the same offer with a better score. What candidates report as harder is the depth: more demanding quantitative and logical items, and coding that expects a genuine working solution with attention to complexity and edge cases rather than pseudocode recognition.

The compensation gap is the reason to care. Reported figures vary by year, batch, degree, and location, but candidates commonly describe the standard entry offer near ₹4.25 LPA and the higher band track near the ₹7 to ₹7.5 LPA region. Treat those as reported ranges, never quote them in an interview as though they were policy, and confirm what applies to your own drive.

The strategic advice is simple: if the higher track is open to you and your coding is reasonable, attempt it, because no negotiation after the fact will close a gap of that size. What you must not do is sacrifice preparation for the standard route in order to chase it.

Answer

The recurring set is narrow enough to drill to reflex: percentages, profit and loss, ratio and proportion, averages, time speed and distance, time and work, simple and compound interest, number systems including HCF, LCM, remainders and divisibility, basic permutation and combination, probability, mensuration, and data interpretation. The skill actually being measured is not mathematics, it is arithmetic economy under a clock. Learn the standard conversions so you stop doing long division: 12.5 percent is one eighth, 16.67 percent is one sixth, 37.5 percent is three eighths.

Use the successive change formula instead of computing two stages separately. Treat total work as the LCM of the individual times so you work in whole units instead of fractions. Back solve from the options whenever the algebra looks longer than the arithmetic.

Memorise squares to thirty, cubes to fifteen, and reciprocals to twenty. The preparation mistake that wastes the most time is solving a thousand random questions. Solve two hundred and record, for every mistake, whether it was a concept gap, a calculation slip, or a misread of the question. Almost everyone finds one category dominating the log, and fixing that category is where the marks actually are.

TIME AND WORK, the LCM method

A finishes a job in 15 days, B in 10 days. Together?

1. Total work = LCM(15, 10) = 30 units
2. A's rate    = 30 / 15 = 2 units per day
3. B's rate    = 30 / 10 = 3 units per day
4. Combined    = 5 units per day
5. Time        = 30 / 5 = 6 days

SUCCESSIVE PERCENTAGE CHANGE

Net = a + b + (a * b / 100)
Up 25 percent then down 20 percent:
25 + (-20) + (25 * -20 / 100) = 25 - 20 - 5 = 0 percent net

SIMPLE vs COMPOUND INTEREST, 2 year gap

Difference = P * (R / 100) ^ 2
P = 20000, R = 10  ->  20000 * 0.01 = 200

AVERAGE shortcut
New average = old average + (entrant value - old average) / new count

Answer

Reasoning is usually the most clearable block for a prepared candidate because the question types are finite. Expect seating arrangements both linear and circular, blood relations, coding and decoding, syllogisms, direction sense, number and alphabet series, statement and conclusion, data sufficiency, and visual or figure based pattern items. The difference between a fast solver and a slow one is diagramming discipline, not intelligence.

For seating arrangements, draw the frame first, fill in only definite information, and mark uncertain positions with an option list instead of committing to a guess you will have to unwind. For blood relations, use a symbol convention you never vary, plus for male, minus for female, a horizontal line for spouse, a vertical for parent to child. For syllogisms, draw Venn diagrams and then test whether the conclusion holds in every valid arrangement rather than only the first one you drew, because testing a single arrangement is the single most common error in this section.

For alphabet series, memorise the position anchors, E is 5, J is 10, O is 15, T is 20, Y is 25, and the questions become arithmetic. Nobody will ask you a syllogism in the interview, so this block is purely a marks opportunity and it is the cheapest section to improve in two weeks.

Answer

They test different things and need different preparation. The verbal ability block is recognition based: reading comprehension, sentence completion, error spotting, para jumbles, synonyms and antonyms, and article and preposition usage, all answered by selecting an option. The separate English communication test is production based, you have to generate correct written English yourself against a word count and a timer.

A candidate can score well on the first and badly on the second, which is exactly what happens to people who prepared with option based practice only. Losses in the verbal block cluster in three predictable places. Articles and prepositions, because those are learnt by exposure rather than by rule and Indian regional English has settled habits the test marks wrong.

Subject verb agreement in long sentences where a clause separates the subject from the verb. And reading comprehension, not because the passages are hard but because candidates read them fully and carefully and then run out of time. Read the questions first, scan for the location, and read closely only around it. For grammar, do not study grammar abstractly; work through error spotting sets and write down the specific rule you broke each time, because your errors will repeat across only eight or ten rules.

Answer

Marking rules have differed across cycles and even across sections within the same assessment, so read the instruction screen on test day rather than assuming. The decision framework is the same either way. If there is no negative marking, never leave a question blank: keep the last thirty seconds of every section free purely to sweep and mark something on everything unattempted.

If there is negative marking, apply elimination discipline, guess only once you have ruled out at least two options because that converts a one in four gamble into a favourable bet, and skip cleanly when you cannot eliminate anything. The costliest mistake is emotional guessing on a hard question you have already spent four minutes on, because the sunk time makes you unwilling to walk away, and that single question then costs you the three easy ones you never reached. Decide your rule before the test and follow it mechanically rather than deciding live.

Also note that most candidates lose more marks to misreading than to guessing, particularly on items asking for the value that is not true, or asking for the remaining quantity when you computed the used quantity. Underline what is actually being asked before you compute.

Answer

The written English test is the most distinctive part of the Capgemini fresher funnel and the part candidates are least ready for. It exists because the verbal section only proves you can recognise correct English, while the job requires you to produce it: status updates to a client, defect descriptions, handover notes, and email that a stakeholder in another country will read without you there to explain it. So the test asks you to type an essay on a given topic, within a stated word count, inside a running timer.

What is being scored is not literary quality. Candidates and evaluators consistently describe the same criteria: whether you addressed the topic given, whether the piece has a visible structure, whether your grammar and spelling hold up, whether sentences are clear rather than ornate, and whether you respected the word count and finished. Two candidates with identical ideas score differently purely on structure and spelling.

The framing to hold in your head is that this is a business writing test wearing an essay costume. Write the way you would write to a client, plain, organised, and finished, not the way you wrote in a school examination. A weak submission is three long unbroken paragraphs of opinion. A strong one is short paragraphs, each doing one job.

Answer

Word counts and durations reported by candidates vary between cycles, commonly falling in the range of a couple of hundred words within a short fixed window, so confirm the exact numbers on your instruction screen and then work to a fixed template rather than improvising. The template that survives a timer is four blocks: a two sentence opening that states your position on the topic, two body paragraphs each carrying one point plus one concrete example, and a two sentence close that restates the position and adds a forward looking line. Allocate the clock before you type: roughly the first tenth of the time planning, the next seven tenths writing, and the final fifth reading back for spelling and sentence breaks.

That final read back is not optional; it is where the cheapest marks in the whole assessment are, because spelling and agreement errors are counted mechanically. Write to the middle of the permitted range rather than to the ceiling, since a piece that runs over and gets cut mid sentence looks unfinished and one that is far under looks thin. Count words in blocks rather than individually, roughly forty words per short paragraph, and keep sentences under about twenty words each because long sentences are where grammar collapses under time pressure.

WORKED ESSAY OUTLINE
Topic: Is remote working good for freshers in the IT industry?

BLOCK 1, OPENING (about 40 words)
  Sentence 1: state the position plainly.
    "Remote working suits experienced professionals far better than it suits
     freshers, and companies should treat the two groups differently."
  Sentence 2: signal the two reasons you will give.

BLOCK 2, BODY POINT ONE (about 60 words)
  Claim: freshers learn by observation, not by documentation.
  Example: a new joiner who overhears a production issue being debugged
           learns more in an hour than a week of reading runbooks.

BLOCK 3, BODY POINT TWO (about 60 words)
  Claim: remote work removes the informal feedback loop.
  Example: a small mistake corrected in passing at a desk becomes a formal
           review comment days later when everyone is remote.

BLOCK 4, CLOSE (about 40 words)
  Restate the position without repeating the wording.
  Add one forward looking line: a hybrid start, remote later.

TIME PLAN on a 20 minute timer
  0 to 2    plan the four blocks in the answer box itself
  2 to 16   write straight through, do not edit while writing
  16 to 20  read back once for spelling, agreement, and paragraph breaks

Answer

Topics are typically general and opinion based rather than technical, drawn from familiar territory such as technology and society, work culture, education, the environment, social media, or a current general interest theme. You are not being tested on domain knowledge, which is exactly why an unfamiliar topic is survivable. The failure is behavioural, not intellectual: candidates freeze, spend six of twenty minutes deciding what they think, and then write in a panic.

Have a recovery procedure ready. First, take a position within thirty seconds and commit to it, because either side scores equally and the marks are in the structure. Second, if you know nothing specific, argue from principles you do have, cost, fairness, time, risk, and learning, all of which apply to almost any topic.

Third, use an example from your own life rather than inventing a statistic, because a fabricated figure reads as false and a personal example never can. A line like 'in my college the same problem appeared when' is always available to you. Fourth, if the topic genuinely allows both sides, say so in the opening, argue the stronger side, concede one point to the other, and close, which is a legitimate and well structured essay. What sinks candidates here is silence, not ignorance.

Answer

The recurring list is short and every item is avoidable. Spelling errors, which are the cheapest marks to lose and the easiest to catch in a read back, with the worst offenders being words you use constantly, receive, definitely, separate, occurred, environment, government, and business. Subject verb agreement over long sentences.

Tense drift, starting in present and sliding into past halfway through. Missing paragraph breaks, which turns a well argued piece into an unreadable block and is scored as poor structure even when the content is fine. Overwriting, meaning long ornate sentences and vocabulary you are not fully in control of, because a misused word is more damaging than a plain one.

Not addressing the actual topic, which happens when a candidate writes the essay they prepared instead of the one they were asked for. Running over the word limit so the piece is truncated mid sentence, or stopping far short so it reads as thin. And SMS or chat register, contractions everywhere, informal openings, or an abbreviation such as 'etc' doing the work that a real example should do. Fix the mechanical ones with a disciplined read back and you will be ahead of most of the pool without writing any better.

Answer

Practise the constraints, not the platform. Set a timer to the duration you expect, open a plain text editor with spell check switched off, pick a topic from a list you did not write, and type an essay to the word count without stopping. Switching spell check off matters more than anything else, because assessment editors usually do not correct you and candidates who always write with a red underline safety net have no idea how many words they actually misspell.

Do this eight to ten times across two weeks and keep every attempt in one folder. After each one, do two things: count the words to see how badly you estimate length, and mark every error in a different colour and tally the error types. Almost everyone finds the same three or four errors repeating, and fixing those specific three is a bigger score improvement than reading a grammar book.

Also practise typing speed if you are below roughly twenty five to thirty words per minute, because below that the timer alone will beat you regardless of how good your English is. Finally, read one well written business article a day, not for content but for sentence length, because imitation of plain professional prose is the fastest route to a clean essay.

Answer

Pseudocode sections are a hallmark of the Capgemini style assessment and they measure something different from a coding round. A coding problem asks whether you can produce a solution. A pseudocode question hands you a small block of language neutral code and asks what it prints, what the value of a variable is after execution, how many times a loop body runs, or which line contains an error.

That is comprehension, and it maps directly to the work: a service delivery engineer spends far more time reading code somebody else wrote than writing new code. The format is usually multiple choice with a short block written in a generic syntax that resembles C, sometimes with explicit declarations such as Integer or String, sometimes with Set for assignment and End For to close a loop. Candidates who only ever write Python are the ones who struggle, because they are not used to declared types, integer division, or explicit loop bounds.

The preparation that works is mechanical: solve fifty of these on paper, tracing every variable in a table, rather than reading the block and guessing the answer from the shape of it. Guessing works on the easy items and fails on exactly the ones that separate candidates.

Answer

The method is a variable trace table and it is worth more than any amount of cleverness. Draw a column for each variable plus a column for the output, then execute one line at a time and write the new value in the next row. Never hold two variables in your head, because that is where every mistake in this section comes from.

Three specific traps recur. Loop bounds: check whether the condition is less than or less than or equal to, because that single character changes the iteration count, and check whether the counter increments by one or by more. Integer division: in a typed pseudocode, seven divided by two is three, not three point five, and questions are built specifically around that truncation.

Operator precedence and the modulo operator, where i mod 2 equals zero is testing evenness and candidates misread it as assignment. In the block below, the accumulator alternates between adding and subtracting, which is designed so that anyone who guesses from the shape of the code answers with the plain sum. The trace takes forty seconds and gets it right.

A weak approach is reading the block twice and choosing the option that feels right. A strong approach is one trace table per question, every time, even when it looks obvious.

PSEUDOCODE

  Function Compute(Integer n)
      Integer i, s
      Set s = 0
      for (i = 1; i <= n; i = i + 1)
          if (i mod 2 == 0)
              Set s = s + i
          else
              Set s = s - i
          end if
      end for
      return s
  End Function

  Print Compute(5)

TRACE TABLE

  step | i | i mod 2 | branch   | s
  =====================================
  init | 1 |    1    | odd      |  0
   1   | 1 |    1    | s = s - 1|  -1
   2   | 2 |    0    | s = s + 2|   1
   3   | 3 |    1    | s = s - 3|  -2
   4   | 4 |    0    | s = s + 4|   2
   5   | 5 |    1    | s = s - 5|  -3

  Loop ends because i becomes 6 and 6 <= 5 is false.

OUTPUT
  -3

WHAT CHANGES THE ANSWER
  i < n instead of i <= n      -> loop runs 4 times, output is 2
  i = i + 2 instead of i + 1   -> only odd i, output is -9

Answer

This family of question separates candidates who understand memory from candidates who have memorised syntax, which is exactly why it appears. When a parameter is passed by reference, typically marked with an ampersand in a C style pseudocode, the function receives an alias for the caller's variable, so any change inside the function is visible outside it. When a parameter is passed by value, the function receives a copy, so changes inside are discarded when the function returns.

The trap is that both variables are modified identically inside the function and both print the same value there, so the line printed inside the function looks like proof that both changed. Only the print after the return reveals the difference. Candidates who trace only the inside of the function answer with the same pair twice.

Two follow ups are common and worth having ready. First, arrays: in C style languages an array parameter decays to a pointer, so array contents behave as if passed by reference even without an explicit marker, which is a favourite second question. Second, Java, where everything is passed by value but object references are themselves the values, meaning you can mutate the object a reference points to but cannot reassign the caller's reference. Being able to state that distinction cleanly is a strong signal in the interview too.

PSEUDOCODE

  Function Modify(Integer &x, Integer y)
      Set x = x + 10
      Set y = y + 10
      Print "inside: ", x, y
  End Function

  Integer p, q
  Set p = 5
  Set q = 5
  Call Modify(p, q)
  Print "outside: ", p, q

TRACE

  1. p = 5, q = 5
  2. Modify is called. x is an ALIAS for p. y is a COPY of q.
  3. x = x + 10  ->  p itself becomes 15
  4. y = y + 10  ->  the local copy becomes 15, q is untouched
  5. Print inside  ->  inside: 15 15
  6. Function returns, the copy y is destroyed
  7. Print outside ->  outside: 15 5

OUTPUT
  inside: 15 15
  outside: 15 5

FOLLOW UP YOU SHOULD EXPECT
  Arrays in C style languages decay to a pointer, so the callee can change
  the caller's array contents even with no reference marker.
  In Java everything is passed by value, but the value of an object variable
  is a reference, so you can mutate the object and cannot reassign the caller.

Answer

The base problem is trivial and that is the point: the panel is testing what you do after the trivial part. Solve it with two pointers moving inward from both ends, which is O(n) time and O(1) extra space and is what they want to see, rather than reversing the string and comparing, which costs O(n) extra space. Then handle the follow ups before they are asked, because that is what earns the marks.

Should case be ignored? Should spaces and punctuation be ignored, so that 'A man, a plan, a canal, Panama' counts as a palindrome? What about an empty string or a single character, both of which are conventionally palindromes but you should state your contract rather than assume.

What about Unicode or digits? Ask the case and punctuation question out loud before you write a line, because asking clarifying questions is itself scored in the technical round. The next follow up in this family is usually checking whether a number is a palindrome without converting it to a string, which you solve by reversing the number arithmetically with modulo and integer division and comparing, and then the interviewer will point out that reversing a large integer can overflow, so reverse only half the digits. Have that answer ready.

PYTHON, two pointers, ignoring case and non alphanumerics

def is_palindrome(s):
    i, j = 0, len(s) - 1
    while i < j:
        while i < j and not s[i].isalnum():
            i += 1
        while i < j and not s[j].isalnum():
            j -= 1
        if s[i].lower() != s[j].lower():
            return False
        i += 1
        j -= 1
    return True

print(is_palindrome("A man, a plan, a canal, Panama"))   # True
print(is_palindrome(""))                                  # True by convention

JAVA, integer palindrome without converting to a string

static boolean isPalindrome(int x) {
    if (x < 0 || (x % 10 == 0 && x != 0)) return false;
    int reversedHalf = 0;
    while (x > reversedHalf) {
        reversedHalf = reversedHalf * 10 + x % 10;
        x /= 10;
    }
    return x == reversedHalf || x == reversedHalf / 10;
}

Time O(n) and O(log n) respectively, extra space O(1) in both.

Answer

This is the SQL question that separates people who have memorised queries from people who understand grouping, because the average has to be computed per department and then compared row by row, which a plain GROUP BY cannot do on its own. There are three good answers and the panel wants you to know more than one. The correlated subquery version is the classic: for each employee row, run a subquery computing the average for that employee's department.

It is readable and portable, and its weakness is that conceptually it evaluates per row, so it can be slow on large tables. The join to a derived table version computes each department average once and joins back, which is usually the better plan on real data and is the answer to give if the interviewer asks about performance. The window function version is the modern one, computing AVG over a partition by department alongside each row without collapsing it, and it generalises immediately to questions like top three earners per department using ROW_NUMBER or DENSE_RANK.

Volunteer two things unprompted: NULL salaries are ignored by AVG, which can quietly skew your comparison, and employees with no department need an explicit decision. A weak answer produces one memorised query and cannot adapt when the interviewer changes it to per department per year.

/* 1. Correlated subquery, the portable classic */
SELECT e.emp_id, e.name, e.dept_id, e.salary
FROM   employee e
WHERE  e.salary > (
         SELECT AVG(e2.salary)
         FROM   employee e2
         WHERE  e2.dept_id = e.dept_id
       );

/* 2. Derived table, each average computed once, usually the better plan */
SELECT e.emp_id, e.name, e.dept_id, e.salary
FROM   employee e
JOIN ( SELECT dept_id, AVG(salary) AS avg_sal
       FROM   employee
       GROUP BY dept_id ) d
  ON   d.dept_id = e.dept_id
WHERE  e.salary > d.avg_sal;

/* 3. Window function, generalises to top N per department */
SELECT emp_id, name, dept_id, salary
FROM ( SELECT emp_id, name, dept_id, salary,
              AVG(salary) OVER (PARTITION BY dept_id) AS avg_sal
       FROM   employee ) t
WHERE salary > avg_sal;

/* Follow up: top 2 earners per department */
SELECT emp_id, name, dept_id, salary
FROM ( SELECT emp_id, name, dept_id, salary,
              DENSE_RANK() OVER (PARTITION BY dept_id ORDER BY salary DESC) AS rnk
       FROM   employee ) t
WHERE rnk <= 2;

Answer

Do not rewrite the algorithm. A solution passing six of ten is almost never conceptually wrong, it is failing one category of input, and your job is to identify which category rather than to start again. Work a fixed checklist in order.

Empty input and single element input. Maximum constraint size, which in C or Java usually means integer overflow and the fix is long instead of int. Negative numbers and zero.

All elements identical. Already sorted and reverse sorted input. Leading, trailing, and repeated whitespace, plus case variation, if it is a string problem.

Then read the verdict type, because it narrows the search before you touch the code: a wrong answer verdict means a logic or edge case issue, a time limit exceeded verdict means your complexity is too high and the logic may be perfectly correct, and a runtime error verdict points you straight at array bounds or division by zero. Also confirm you are reading input in exactly the specified format and printing exactly the specified output, with no prompt strings such as 'Enter a number', because an automated checker counts those as wrong output and that alone has cost candidates entire sections. A weak answer is 'I would check my logic again', which says nothing. A strong one names the checklist and explains that the verdict tells you where to look.

Answer

Name exactly one language and commit to it. This is not a modesty question, it is the panel choosing which door to open, and a candidate who answers 'C, C plus plus, Java, Python and a bit of JavaScript' has just invited questions in all five and will be examined on their weakest. Say one language, add one sentence about where you used it, and stop talking.

For example, 'Java, I have used it for two years including my final year project and my data structures coursework'. The follow ups are predictable per language and you should prepare that specific set. If you name Java, expect object oriented concepts, abstract class against interface, the collections framework, string immutability and the string pool, exception handling with checked against unchecked, and threads at a surface level.

If you name Python, expect list against tuple, mutable default arguments, dictionaries, comprehensions, shallow against deep copy, and generators. If you name C, expect pointers, memory allocation, structures, and string handling. Depth in one language beats shallowness in four every time.

If you genuinely work across two, name the primary and mention the second only as an aside. A weak answer hedges. A strong one commits and then grounds the language in something you built, which quietly steers the conversation toward your project where you are strongest.

Answer

Panels have heard the car and animal examples several hundred times and both signal that a candidate memorised rather than built. Ground every pillar in your own project. Encapsulation is bundling state with the methods that act on it and controlling access, so you might say your payment class keeps card details as private fields and exposes only a process method, which means no other class can leave the object half valid.

Abstraction is exposing what an object does while hiding how, so a report generator interface with a single generate method lets the caller work without knowing whether the output is a PDF or a spreadsheet. Inheritance is reuse with specialisation, a base user class with admin and student subclasses adding their own permissions, and you should immediately add the caveat that inheritance is overused and composition is usually the better default, which signals maturity rather than recitation. Polymorphism is one interface with many implementations, resolved at compile time through overloading or at runtime through overriding, and you must be ready to write both on the spot because 'give me an example of runtime polymorphism from your own code' is the guaranteed follow up.

A weak answer defines all four in thirty seconds and cannot produce a single line of code. A strong one gives one concrete example per pillar and volunteers the composition caveat.

Answer

Structurally, a class can implement many interfaces but extend only one abstract class, and that constraint alone often decides the design. An abstract class can hold instance state, constructors, and non public members. An interface historically could not, although modern Java allows default methods, static methods, and private methods inside interfaces, plus constant fields.

The decision rule worth stating is about relationships rather than syntax. Use an abstract class when the subtypes are genuinely the same kind of thing and share implementation, an is a relationship with shared state, for example an abstract Vehicle holding a registration number and a common start sequence. Use an interface when you are describing a capability that otherwise unrelated types can have, a can do relationship, for example Printable or Comparable, which a report, an invoice, and an identity badge might all implement without being related to each other.

The trap follow up is 'since Java 8 added default methods, why do we still need abstract classes'. The answer is state and construction: an interface still cannot hold instance fields or run a constructor, so shared mutable state belongs in an abstract class. A weak answer lists syntax differences only. A strong one gives the is a against can do rule and disposes of the default methods follow up before being led there.

/* Abstract class: shared state plus a constructor, an is a relationship */
abstract class Vehicle {
    protected final String regNumber;
    Vehicle(String regNumber) { this.regNumber = regNumber; }
    void start() { System.out.println(regNumber + " starting"); }
    abstract int wheels();
}

/* Interface: a capability, unrelated types can implement it */
interface Printable {
    String render();
    default String header() { return "=== report ==="; }
}

class Truck extends Vehicle implements Printable {
    Truck(String r) { super(r); }
    int wheels() { return 6; }
    public String render() { return "Truck " + regNumber; }
}

/* Why the abstract class still exists after Java 8 default methods:
   an interface cannot declare regNumber as instance state,
   and it cannot run a constructor to initialise it. */

Answer

Normalisation is the one database topic where a candidate either understands the mechanism or is reciting, and the difference shows within thirty seconds. Work through an actual table instead of defining the forms. First normal form requires atomic values and no repeating groups, so a subjects column holding 'maths, physics, chemistry' in one cell violates it and those become separate rows.

Second normal form applies only when the primary key is composite and requires every non key attribute to depend on the whole key, so in a table keyed on student id plus course id, the student name depends only on the student id and must move out to a student table. Third normal form removes transitive dependency, where a non key attribute depends on another non key attribute, so if the student table stores department id and department head, the head depends on the department and belongs in a department table. Then say the thing most candidates never say: normalisation is not free, it adds joins, and in reporting and analytics workloads you deliberately denormalise for read speed.

Naming that tradeoff shows you have thought about production rather than exams. A weak answer recites the three forms with no example and cannot answer 'when would you not normalise'. A strong one draws the tables and volunteers the tradeoff.

/* Violates 1NF: repeating group inside one column */
/* student_id | name | subjects                        */
/* 101        | Asha | maths, physics, chemistry       */

/* 1NF: atomic values, one row per enrolment */
CREATE TABLE enrolment (
  student_id INT,
  course_id  INT,
  PRIMARY KEY (student_id, course_id)
);

/* Violates 2NF: student_name depends on only part of the composite key */
/* enrolment(student_id, course_id, student_name, course_fee)           */

/* 2NF: split the partial dependencies out */
CREATE TABLE student (
  student_id INT PRIMARY KEY,
  student_name VARCHAR(80),
  dept_id INT
);
CREATE TABLE course (
  course_id INT PRIMARY KEY,
  course_fee DECIMAL(10,2)
);

/* Violates 3NF: dept_head depends on dept_id, not on student_id */
/* student(student_id, student_name, dept_id, dept_head)         */

/* 3NF: remove the transitive dependency */
CREATE TABLE department (
  dept_id INT PRIMARY KEY,
  dept_head VARCHAR(80)
);

Answer

Definitions alone will not clear this, because the panel usually draws two small tables and asks how many rows each join returns, which is where memorised answers collapse. INNER JOIN returns only matching pairs. LEFT JOIN returns every left row plus its matches, with NULLs where there is no match.

RIGHT JOIN is the mirror. FULL OUTER JOIN returns everything from both sides with NULLs where unmatched, and it is worth volunteering that MySQL does not support it directly so you emulate it with a UNION of the left and right joins. CROSS JOIN returns the Cartesian product.

SELF JOIN is a table joined to itself, typically for an employee and manager hierarchy, which is a common follow up in its own right. Two things trip candidates up and you should raise both before you are asked. Fan out: if the right table has three rows matching one left row, the join returns three rows and not one, and that silent multiplication is what doubles totals in reports.

And the filter placement trap: a condition on the right table placed in WHERE rather than in ON converts a LEFT JOIN into an effective INNER JOIN, because the NULL rows fail the predicate. A candidate who explains fan out unprompted is answering the question after the question.

/* emp:  3 rows, dept_id values 10, 10, NULL */
/* dept: 2 rows, ids 10 and 20                */

SELECT * FROM emp e INNER JOIN dept d ON e.dept_id = d.id;   /* 2 rows */
SELECT * FROM emp e LEFT  JOIN dept d ON e.dept_id = d.id;   /* 3 rows */
SELECT * FROM emp e RIGHT JOIN dept d ON e.dept_id = d.id;   /* 3 rows */
SELECT * FROM emp e CROSS JOIN dept d;                       /* 6 rows */

/* SELF JOIN: employee to manager */
SELECT e.name AS employee, m.name AS manager
FROM   emp e
LEFT JOIN emp m ON e.manager_id = m.id;

/* The trap: this LEFT JOIN behaves like an INNER JOIN */
SELECT * FROM emp e
LEFT JOIN dept d ON e.dept_id = d.id
WHERE d.name = 'Sales';

/* The fix: move the filter into the ON clause */
SELECT * FROM emp e
LEFT JOIN dept d ON e.dept_id = d.id AND d.name = 'Sales';

Answer

This carries the highest weight in the technical round because it is the only part of the interview where you control the material, and panels use it to test whether the project is genuinely yours. Structure the answer in four beats and keep it under ninety seconds before pausing for questions. What problem it solved and for whom.

What you personally built, using 'I' for your parts and being honest about the team split. The stack, plus one design decision you made and the alternative you rejected and why. And one thing that went wrong and how you fixed it.

That fourth beat separates candidates, because a project with no problems in it was not built. Expect the panel to drill into every technology you name, so never list a framework you cannot survive three questions about. If you used a database, expect to be asked for your schema and to justify a table design.

If you called an API, expect to be asked how you handled failures and timeouts. If you claim machine learning, expect questions about dataset size and your evaluation metric, and be ready to say why accuracy was or was not the right metric. A weak answer is a feature tour with no decisions in it. A strong one is a decision narrative where you can defend every choice you made.

Answer

Nobody expects a fresher to be a cloud architect, and claiming to be one is a fast way to get dismantled. What is expected is vocabulary you can use correctly plus one or two things you have actually touched. Be able to explain the three service models in your own words with an example each: infrastructure as a service where you rent a virtual machine and manage the operating system yourself, platform as a service where you deploy an application and the platform manages the runtime, and software as a service where you consume a finished application.

Know what a region and an availability zone are and why deploying across zones matters. Know the difference between vertical and horizontal scaling and what a load balancer does. Know that object storage such as a bucket is not a file system and why that matters for application design.

Be able to say one sentence about what containers solve and what an orchestrator adds on top. If you have deployed anything anywhere, even a college project on a free tier virtual machine, lead with that, because one real deployment beats a list of service names every time. The honest and strong closing line is that you have used a small set at a basic level and you know the vocabulary well enough to learn fast on a project, which is exactly what a fresher is expected to say.

Answer

The depth is conceptual and the sampling is wide, so prepare one clear sentence plus one example per topic rather than paragraphs on a few. On networking: the OSI layers with one protocol named per layer, TCP against UDP where the honest answer is reliability and ordering against latency and overhead with a real example each, file transfer against live video. What happens when you type a URL and press enter, which is the single most asked networking question anywhere and should be delivered as a clean chain, DNS resolution, TCP handshake, TLS handshake, HTTP request, server response, browser rendering.

IP addressing basics, public against private addresses, what a port is, what DNS does, and the HTTP status code families with one example each. On operating systems: process against thread, where the useful answer is that threads share the address space and processes do not, which is exactly why threads are cheaper and exactly why they need synchronisation. Deadlock with its four necessary conditions and how breaking any one prevents it.

Virtual memory and paging at a conceptual level, plus thrashing. Scheduling with at least round robin and first come first served, and the idea of starvation. A weak answer is a definition with no example attached. A strong one attaches a concrete situation to every concept.

Answer

This is the great filter of the managerial round and it helps to understand the honest context. A large services business runs on long lived client systems, and a meaningful share of fresher allocation goes into production support, maintenance, testing, package implementation, and older technology stacks. Those projects fund the business and the work is genuinely skilled.

What the panel is checking is whether you will become a retention risk in month four. The strong answer accepts the allocation without pretending to be indifferent: yes, you will work on whatever you are allocated, plus a real reason rather than a compliant one, for example that in a support role you learn how a production system actually behaves under load and how a business genuinely uses it, which is knowledge you cannot get from a greenfield build. Then, and only then, name your interest area once as a longer term aspiration and ask how internal movement works after the first project.

Asking that reads as planning rather than resistance. A weak answer is 'I only want to work in AI and full stack', which is the single fastest way to lose an offer at this stage. Another weak answer is a flat 'anything is fine' with no reasoning behind it, which reads as compliance rather than commitment and is remembered as neither.

Answer

The panel is testing whether you can disagree without escalating, because delivery work is full of people who disagree and still have to ship on a date. Use a real, small, specific disagreement rather than a dramatic one, and structure it as situation, what the disagreement actually was, what you did, and the outcome including what you learnt. The most common failure is choosing a conflict where you were entirely right and the other person entirely wrong, because that tells the panel nothing except that you keep score.

Pick one where the other person had a legitimate point. A good shape is two of you disagreeing on whether to spend the last week adding a feature or stabilising what already existed, you going to the evaluation criteria rather than arguing, and the criteria settling it without either of you conceding on ego. That shows you resolve disagreements with evidence rather than volume or hierarchy.

Avoid three things: claiming you have never had a disagreement, which is not believable, describing one you resolved by silently doing all the work yourself, which signals you cannot delegate, and naming a teammate in a way that sounds like blame. Close with the specific habit you changed afterwards, because the learning line is what actually gets written on the scorecard.

SAMPLE ANSWER

"In our final year project two of us disagreed in the last week. I wanted to
add a dashboard screen, my teammate wanted to fix the defects we already knew
about. Both positions were reasonable. Rather than argue it out, I pulled up
the evaluation rubric and we saw that demo stability carried more weight than
feature count. We spent the week fixing defects, the demo ran without a crash,
and I have used that approach since. When two people disagree, go and find the
criteria instead of debating opinions."

WHY THIS WORKS
- The other person had a legitimate point, so it is a genuine disagreement
- Resolved with evidence rather than seniority or volume
- Concrete verifiable outcome, the demo did not crash
- Ends with a transferable habit, which is what gets scored

Answer

This is a stress question and the content of your answer matters less than your composure, because the panel is deliberately applying pressure to see whether you become defensive, deflate, or hold your position calmly. Client facing work involves being challenged by people with more authority than you, and that is exactly what is being simulated. Do not argue with the score.

Do not blame the test, your internet connection, your college, or the section timing. Do not collapse into apology either. Acknowledge it in one clean sentence, then redirect to evidence they can verify: the project you built and can demonstrate, the section you did clear, an internship, a certification you completed, or your performance in this round.

Close by inviting them to test you on something, which converts the exchange from defence into an offer. Three things to avoid: over explaining, which reads as anxiety, comparing yourself to other candidates, and going silent. If you genuinely have no counter evidence, say what you have done since the test and what you are doing now, because effort in progress still beats an excuse. What is being scored here is temperament under challenge, and you can lose the point even with good content if you deliver it defensively or too fast.

Answer

Situational questions like this appear because escalation handling is a daily reality in delivery and because the answer reveals whether you understand your own authority level. The structure that works has four parts. Listen and let the client finish without interrupting or defending, because most escalations contain a factual complaint wrapped in frustration and interrupting guarantees you address the frustration instead of the fact.

Acknowledge the impact in their terms, not yours, which means saying what it cost them rather than what went wrong technically. Separate what you can commit to from what you cannot: as a junior you can commit to investigating and reverting with a specific update time, and you must not commit to a fix date, a root cause, or a compensation you have no authority over. And escalate internally immediately and transparently to your lead, before the client escalates it above you, because a lead who hears it from their own team first can act, whereas a lead who hears it from the client is already behind.

A weak answer promises the client an immediate fix to make the call end, which creates a second failure when the promise breaks. Another weak answer is going silent and hoping a senior speaks. Composure plus an honest boundary is the whole of what is being scored.

Answer

Ninety seconds, four beats, and no life story. The panel is not collecting information, they already have your resume in front of them. They are assessing spoken fluency, structure, and confidence, and they are deciding which thread to pull next, which makes this your one chance to choose the topic of the next five minutes.

Beat one: who you are academically in one line, degree, branch, college, graduating year. Beat two: your strongest technical grounding named specifically, one language and one area. Beat three: one project or internship with an actual result, quantified if you can, and this is the thread you want them to pull.

Beat four: why you are here, connecting what you have done to the role. The common failures are consistent: starting with where you were born, listing every family member, listing every subject you studied, running to four minutes, or reciting something obviously memorised in a flat monotone. Practise it aloud until it is fluent but not word perfect, because word perfect sounds rehearsed and a panel that senses recitation will interrupt you.

Do not say anything you cannot defend, because every noun in this answer is a potential question. End cleanly with a full stop rather than trailing off, since a confident close is disproportionately memorable.

SAMPLE ANSWER, about 90 seconds

"I am Sneha Kulkarni, a final year Computer Engineering student at a college
in Pune, graduating in 2026 with 8.2 CGPA.

My strongest area is Java, along with SQL and database design, which I built
up across coursework and two projects.

The project I am proudest of is a hostel mess management system I built with
three teammates. I owned the backend and the database schema. Our first
version recalculated every student's monthly bill on each page load and the
page took several seconds, so I moved the calculation to a nightly job and
stored the result, which brought the page down to under a second and let all
six hundred students use it during the billing week.

I am interested in Capgemini because I want to work on systems that real
users depend on daily, and the range of client industries here means I would
see problem types I could not see in a single product company."

WHY THIS WORKS
- Four clean beats, under ninety seconds, no family history
- Every claim is defensible under follow up questioning
- The project beat has a specific problem and a specific measurable fix
- Closes with a reason about the work, not about the brand

Answer

This is not conversational, it is a filter, and a hesitant answer here is frequently the reason a candidate who cleared every technical round does not receive an offer. Delivery centres are spread across cities including Mumbai, Pune, Bengaluru, Chennai, Hyderabad, Noida, Kolkata, and others, and allocation follows project demand rather than candidate preference, so the business cannot promise your city. The strong answer is an unqualified yes followed by one sentence of substance: yes, you are willing to relocate, you have thought about it practically, and you would prefer a particular location if there is a choice but it is not a condition.

Never say 'yes but not the north' or 'only if it is near home', because the panel writes that down verbatim. If you have a genuine constraint, for example caring for a parent with a medical condition, state it once, calmly and factually, and immediately follow with what you can do rather than what you cannot. Do not manufacture a constraint to look serious, and do not say yes while your face says no, because managerial and HR panels are specifically trained to read that gap. What they are really probing is whether you will accept an allocation without escalating three weeks after joining, which is a real and costly operational problem for them.

SAMPLE ANSWER

"Yes, I am completely open to relocating. I have already discussed it at home
and there is no constraint from my side. If there is a choice I would find
Pune or Mumbai easiest for family reasons, but that is a preference and not a
condition. Wherever I am allocated, I will move and I will be there on the
joining date."

WHY THIS WORKS
- The unqualified yes comes first, before any nuance at all
- Shows the decision was actually discussed, not improvised in the room
- Names a preference without converting it into a condition
- Closes with a commitment sentence, which is the line that gets recorded

Answer

Global delivery means projects run to European or American business hours and support projects run rosters that include nights and weekends. The panel needs to know now rather than after you have been staffed onto a project. Answer yes if it is true, and answer with specifics so it does not sound like a reflex: say you understand shifts may rotate, that you are comfortable with it, and that you have thought about the practical side, the commute, transport arrangements, and sleep.

If you have a genuine constraint, be straight about it and offer what you can do instead, for example that evening overlap hours covering a European client work comfortably for you. Do not give a false yes, because the cost of that lands on your team within weeks and damages you far more than a truthful no would have. What a strong candidate adds is the reason overlap hours exist, which is that handovers with the client team happen live rather than over email, so you are not simply accepting an inconvenience, you understand the delivery logic behind it.

A weak answer is 'I can adjust', which is vague and reads as reluctance. Another weak move is asking about the shift allowance in the same breath, save all compensation questions for the end of the HR round.

Answer

The panel knows you applied broadly and is not offended by it, so do not pretend this is your only application. What they are testing is whether you know anything specific or whether you are reciting a template that would work for any employer. The failure mode is generic praise, a global company, a great brand, good learning opportunities, which is what most of the pool says and which panels have stopped hearing.

Build the answer from three specifics. One about the work: the range of client industries and delivery types, which means the systems you touch vary across your career rather than being one product forever. One about your own path: what you want your first two years to look like and why a structured training and delivery environment fits that better than a small team where you would be the only person who knows anything.

One about something you actually researched, a service line, a technology area, or the graduate training programme. If asked directly whether you applied elsewhere, say yes and name what would make you choose this offer. Honesty scores better than a transparently false exclusivity claim, and panels are extremely good at spotting the latter because they hear it thirty times a day.

SAMPLE ANSWER

"I did apply to several companies and I will be honest about that. The reason
Capgemini is where I want to start is the range of work. In one organisation
there is banking, manufacturing, energy, and public sector delivery going on,
so over my first few years I would see problem types I would not see inside a
single product company.

The second reason is the structured start. I know I will not be productive on
day one. A place with a real training programme and experienced people on the
delivery floor is a better foundation for me than a small team where I would
mostly be guessing and nobody would have time to correct me."

WHY THIS WORKS
- Honest about applying elsewhere, which the panel already assumed anyway
- Two specific reasons, neither of which is generic brand praise
- Self aware about being a fresher without being self deprecating
- Contains nothing a follow up question could catch you out on

Answer

Indian IT services employers commonly attach a service agreement to fresher offers, meaning a contractual commitment to remain with the employer for a stated period after joining, usually counted either from the date of joining or from the completion of initial training, with a stated liability if you leave earlier. Durations and amounts vary by year, band, and role, and terms change between cycles, so never rely on what a senior told you about their batch. Read your own document.

Before you sign, check these specifically: the exact duration and whether the clock starts at joining or at training completion, the exact payable amount and whether it reduces on a pro rata basis as you serve, what triggers the liability, whether resignation during training is treated differently from resignation afterwards, what happens if the company terminates you rather than you resigning, whether any original certificates are to be submitted, and the notice period, which is a separate clause from the service agreement and is what governs your exit once the commitment period is over. Ask the HR contact in writing over email rather than verbally so you have the answer on record. If you are asked in an interview whether you are comfortable with a service agreement, answer yes if you are, and it is entirely professional to add that you would like to read the terms.

Answer

Background verification is usually run through a third party agency and checks identity, address, education, prior employment, and in many cases a criminal record check. Candidates worry about the wrong things. Checks almost never fail because of a modest CGPA or a gap year you declared honestly.

They fail because of discrepancies between what you stated and what the record says. The recurring causes are employment dates that do not match what a previous employer confirms, a job title inflated on the resume, a salary figure in the application that payslips do not support, an employment period left off the resume entirely, a degree year or percentage that does not match the university record, and an experience letter from a company that cannot be contacted or will not confirm you. For freshers the common problem is an internship claimed as employment, or a certificate from an institute that cannot verify it.

Two rules follow. Declare everything, including short stints you would rather forget, because an omission discovered later is treated as concealment and is far more serious than the thing itself. And keep your documents in one folder ready: mark sheets and degree certificate, offer and relieving letters, payslips, PAN and Aadhaar, and address proof. If you know of a genuine discrepancy, disclose it to HR proactively, because a disclosed issue is manageable and a discovered one usually is not.

Answer

Campus and off campus fresher offers are made against a demand forecast, and actual onboarding happens when project demand and training capacity line up, which is why candidates frequently wait months between the offer letter and a joining date, sometimes considerably longer in a slow demand cycle. This is normal across Indian IT services and is not by itself a sign that your offer is at risk, but it creates real decisions. Practically: stay reachable, keep the email address on your offer active and check the spam folder, and respond quickly to every communication, because non response is one of the very few things that genuinely puts an offer at risk.

Complete any pre joining learning assigned to you, since completion is often tracked. Keep preparing technically instead of going idle, because assessments follow after you join. The initial training programme after joining typically covers programming fundamentals, a technology stack, database work, and professional and communication skills over several weeks with periodic assessments, and performance there commonly influences your project allocation.

Repeated assessment failure has real consequences that vary by policy and year. The failure pattern is almost always the same: candidates who have just cleared a hard selection process relax for three weeks and then discover the assessment is genuinely demanding. Ask what the retake policy is at the start rather than after you need it.