Technical Interview Questions for Freshers (India, 2026)

Fresher technical interviews in India are more predictable than they feel. Whether it is TCS, Infosys, a product startup or a GCC, the interviewer is working from the same small map: your language basics, one data-structures core, your projects, and - the part nobody warns you about - your own resume. Most rejections happen not on hard problems but on fundamentals the candidate assumed were "too basic to ask".

This guide covers the questions that actually get asked, organised by topic, with what a good answer looks like and how deep the follow-ups usually go. It will not replace practice - but it will stop you from preparing the wrong things.

How fresher technical rounds are structured

Services companies (TCS, Infosys, Wipro, Cognizant, Accenture) run a lighter technical round: fifteen to thirty minutes, fundamentals plus one coding or pseudo-code question, heavy reliance on your resume. Product companies and GCCs run deeper: one or two full technical rounds with DSA problems on a shared editor, plus project grilling. Startups mix both and add "can you start Monday" energy. Know which game you are playing; the preparation overlap is large but the depth bar is not.

One universal rule: everything on your resume is fair game. Every skill listed, every project bullet, every technology named. Interviewers pick interview questions off resumes the way you pick items off a menu. If your resume needs work before these questions matter, the software engineer resume format and fresher resume guides are the starting point.

Language fundamentals (everyone asks)

Pick ONE language and know it properly - Java, Python, C++ or JavaScript. "I know four languages" with shallow answers loses to "I know Java well" every time.

• What is the difference between a list and an array? When would you use each?
• Explain OOP: inheritance, polymorphism, encapsulation, abstraction - with a real example, not a textbook definition.
• What is the difference between == and .equals() (Java) or is and == (Python)?
• How does memory management work - garbage collection in Java/Python, pointers if you claim C++.
• What are exceptions and how do you handle them? What is finally for?
• Java-specific: String vs StringBuilder, HashMap internals, interface vs abstract class. Python-specific: list vs tuple, dictionary internals, generators, GIL basics.

The follow-up pattern is always the same: definition, then "give me an example", then "what happens if...". Prepare each answer three levels deep, not one.

Data structures and algorithms (the coding core)

Services companies ask one easy question; product companies ask one or two medium ones. The repeating set:

• Arrays/strings: reverse a string, find duplicates, two-sum, check palindrome, move zeroes.
• Linked lists: reverse a linked list, detect a cycle - classics that refuse to die.
• Hashing: first non-repeating character, frequency counts, pair-with-given-sum.
• Stacks/queues: balanced parentheses, implement a queue with stacks.
• Recursion: factorial, Fibonacci (and why naive recursion is slow - say "overlapping subproblems" and mention memoisation).
• Sorting/searching: binary search (write it bug-free, including the mid calculation), explain quicksort vs mergesort trade-offs.
• Trees (product companies): traversals, height, level-order with a queue.

How you answer matters as much as what you answer. State the brute force first, name its complexity, then improve. Think out loud - interviewers score the reasoning, and a silent perfect answer scores lower than a talked-through good one. If you blank, say what you are considering rather than freezing; partial credit in interviews is real.

DBMS and SQL (asked almost everywhere)

• Write a query: second-highest salary; employees earning more than their managers; count per department with GROUP BY.
• INNER JOIN vs LEFT JOIN - with an example where they return different row counts.
• What is a primary key, foreign key, unique constraint?
• Explain normalisation - 1NF to 3NF with a quick example table.
• What is an index, and when does an index NOT get used?
• DELETE vs TRUNCATE vs DROP.
• Product companies add: ACID properties, transactions, and "why NoSQL" at a conceptual level.

If SQL appears anywhere on your resume, expect to write at least one query on paper or a shared doc. The second-highest-salary question alone appears in an absurd share of fresher interviews - know two ways to write it.

OS, networks and system basics (selective but brutal)

Services companies ask one or two; product companies and roles touching infrastructure go deeper:

• Process vs thread. What is a deadlock and its four conditions?
• What happens when you type a URL into a browser - the single most common "open" question. Walk it: DNS, TCP handshake, TLS, HTTP, server, render.
• HTTP vs HTTPS, GET vs POST, status codes (200, 301, 404, 500).
• What is an API? REST vs SOAP at a concept level.
• OS scheduling basics, paging vs segmentation (theory courses come back to collect their dues).

Project grilling (where freshers actually lose)

Your final-year project or internship will take ten to twenty minutes of the round. The questions form a fixed pattern:

1. "Explain your project" - have a crisp 60-second version: problem, your solution, stack, outcome. The STAR method shapes this answer well.
2. "Why this stack?" - any honest reason beats "that is what the tutorial used".
3. "What was the hardest bug or decision?" - prepare one specific story with details only the builder would know.
4. "What would you do differently now?" - shows growth; never say "nothing".
5. "Show me" - some interviewers ask you to open the code or draw the architecture. If it is a group project, know exactly which parts YOU built, because they will ask.

The project section is the one place a fresher can sound like an engineer instead of an examinee. Prepare it twice as hard as the theory.

The "resume trap" questions

These are not technical questions; they are honesty checks wearing technical clothes. You wrote "machine learning" because your project used a library - here comes "explain how the model works". You wrote "AWS" because you deployed once - here comes "what is a security group". The rule that saves you: claim only what you can discuss for five minutes. Our HR questions guide covers the non-technical half of the same honesty test.

A two-week preparation plan

1. Days 1-3: one language's fundamentals, three levels deep. Write answers by hand once.
2. Days 4-8: two DSA problems a day from the list above - solve aloud, timed, then check complexity.
3. Days 9-10: SQL - the five query patterns above, written from memory.
4. Days 11-12: your projects - the 60-second pitch, the hard-bug story, the architecture sketch.
5. Day 13: mock interview with a friend or a practice tool; the first time you answer aloud should not be the real round. Running it in CheatCode gives you the grilling without the stakes.
6. Day 14: rest, documents, logistics. Cramming the night before costs more in clarity than it adds in coverage.

Frequently asked questions

Which language should I prepare for fresher interviews?

The one on your resume that you know deepest - for most Indian freshers that is Java or Python. Java is the safer default for services companies and GCCs; Python suits data and startup roles. What loses marks is switching mid-interview or claiming four languages and surviving none of the follow-ups. Depth in one, working familiarity in a second, honesty about the rest.

Are DSA questions really asked in services-company interviews?

Yes, but easy ones - reverse a string, find duplicates, Fibonacci. The bar is "can this person be trained to code", not competitive programming. Product companies and GCCs set the higher bar: one or two medium LeetCode-style problems with edge-case discussion. Calibrate to the company list you are actually applying to.

What if I do not know an answer?

Say so cleanly, then show your thinking: "I have not worked with that directly - my understanding is X, is that the right direction?" Interviewers forgive gaps; they do not forgive confident bluffing, because it predicts an employee who hides problems. One honest "I do not know" per interview costs nothing; three suggest the resume overstated things.

How much do academic subjects like OS and DBMS matter?

For services companies, one or two conceptual questions each - definitions with a real example. For product companies, DBMS and OS can decide the round, especially for roles touching data or infrastructure. If your degree is non-CS, expect "how did you cover these?" - a self-study story with a project attached answers it well.

Should I prepare company-specific questions?

Yes, in the final days before a specific interview. Patterns repeat heavily at scale hirers - the same TCS or Infosys question sets circulate every season. Our company guides - TCS, Infosys, Accenture, Wipro - collect those patterns. But company-specific prep sits on top of fundamentals, never instead of them.