Interview experience
Barclays Interview Questions and Answers (2026)
Overview
Section titled “Overview”Barclays’ India campus process (via BGSC) runs an Online Assessment, one or two technical interviews, and an HR round built around the RISES values framework - a lighter-weight pipeline than the UK graduate route’s full Assessment Centre.
Barclays interview process at a glance
Section titled “Barclays interview process at a glance”| Round | Duration | What it tests |
|---|---|---|
| Online Assessment | ~90 min | CN/OS/DBMS MCQs, 1 SQL query, 1 medium-hard DSA problem |
| Technical Interview 1 | ~60 min | Project deep-dive, OOP, programming fundamentals, DB concepts |
| Technical Interview 2 (some drives) | 45-60 min | Deeper technical + track (frontend/backend/DB) fit |
| HR Round | ~30 min | RISES values, motivation, logistics |
Online Assessment
Section titled “Online Assessment”Around 90 minutes: MCQs on Computer Networks, OS, and DBMS, plus one SQL query you write directly and one DSA problem at medium-to-hard difficulty. This is the main filter - one reported 2025 drive shortlisted 50 of the initial applicant pool.
Common questions
- Computer Networks MCQs - OSI/TCP-IP layers, protocols
- OS MCQs - process scheduling, memory management
- DBMS MCQs plus a live SQL query (joins, aggregates)
- One medium-to-hard DSA problem
Technical Interview 1
Section titled “Technical Interview 1”Roughly 25-30 minutes on your internship or academic projects - technologies used, why you chose them, and challenges faced - followed by OOP principles, programming fundamentals, and database concepts.
Common questions
- Walk through your internship/academic project in detail
- OOP principles - inheritance, polymorphism, encapsulation with real examples
- Java or C++ fundamentals
- Database concepts - normalization, transactions, indexing
Technical Interview 2 (some drives)
Section titled “Technical Interview 2 (some drives)”When present, this round probes technical depth further and checks fit for a specific track.
Common questions
- Deeper follow-up on a data structure or algorithm from Round 1
- Do you prefer frontend, backend, or database work - and why?
- A second, harder coding or design question
- Trade-offs in your project’s architecture
Full round-by-round narratives are on the Barclays interview experience page.
HR Round
Section titled “HR Round”Roughly 30 minutes on resume, extracurriculars, and situational questions, explicitly mapped to Barclays’ RISES values framework.
Common questions
- Tell me about yourself?
- Why Barclays?
- Which RISES value (Respect, Integrity, Service, Excellence, Stewardship) resonates most with you, and why?
- Describe a time you disagreed with a team but implemented the decision anyway
- Do you prefer frontend, backend, or database roles?
Sample answer frameworks for each of these are on the Barclays HR interview questions page.
The Barclays Global Service Centre (BGSC) in India
Section titled “The Barclays Global Service Centre (BGSC) in India”Barclays’ India technology hiring runs through the Barclays Global Service Centre, spread across four sites with genuinely different functional specialisations rather than being interchangeable back offices: Pune (1.1 million sq ft, Barclays’ largest site outside London) leads engineering, analytics, and shared services; Noida focuses on digital engineering and fintech solutions; Chennai handles back-office operations and platform support; and Mumbai covers investment banking and corporate finance technology. Which BGSC site you’re hired into can meaningfully affect the kind of engineering work you do day to day.
Common technical interview questions and answers
Section titled “Common technical interview questions and answers”Q: Explain the OSI model and where common protocols sit.
OSI has seven layers: Physical (bits on the wire), Data Link (framing, MAC addressing, error detection - Ethernet, ARP), Network (logical addressing and routing - IP, ICMP), Transport (end-to-end delivery - TCP, UDP), Session, Presentation (encryption and encoding - TLS sits here in the OSI mapping), and Application (HTTP, DNS, SMTP). The TCP/IP model collapses this into four: Link, Internet, Transport, and Application. The distinction interviewers usually chase is switch versus router: a switch forwards frames using MAC addresses at Layer 2, while a router forwards packets using IP addresses at Layer 3 and separates broadcast domains.
Q: What is the difference between a process and a thread, and what does a context switch cost?
A process has its own address space, file descriptors, and page table, so processes are isolated from one another; threads live inside a process and share the heap, globals, and open files while keeping their own stack, registers, and program counter. That makes thread creation and switching much cheaper, but it also means threads need synchronisation - mutexes, semaphores, or atomics - because they touch shared memory. A context switch between processes must save and restore registers, switch the page table, and typically flush or partially invalidate the TLB, which is the expensive part; a switch between threads of the same process skips the address-space swap. This is why a web server handling many concurrent connections uses a thread pool or async I/O rather than one process per request.
Q: Write a SQL query to find every department whose average salary exceeds the company-wide average.
Use GROUP BY with HAVING and a scalar subquery: SELECT d.dept_name, AVG(e.salary) AS avg_salary FROM Employees e JOIN Departments d ON e.dept_id = d.dept_id GROUP BY d.dept_name HAVING AVG(e.salary) > (SELECT AVG(salary) FROM Employees); The key point interviewers listen for is why the filter is in HAVING and not WHERE: WHERE is evaluated on individual rows before grouping, while HAVING is evaluated on the aggregated groups, so an aggregate function is only legal in HAVING. The inner subquery is uncorrelated and runs once, so the cost is one scan for the average plus the grouped scan.
Q: What is the difference between a clustered and a non-clustered index?
A clustered index determines the physical order of rows in the table, so there can be only one per table - in most engines the primary key is the clustered index and the table is effectively stored as that B+ tree with the full row in the leaves. A non-clustered index is a separate structure whose leaves hold the indexed columns plus a pointer or the clustered key, so reading extra columns needs a second lookup into the base table, often called a key lookup or bookmark lookup. You avoid that by making the index covering - including the extra columns the query selects - so the query is answered entirely from the index. Indexes speed reads but slow inserts, updates, and deletes because every index must be maintained, which is why you do not index every column.
Q: Explain compile-time versus run-time polymorphism in Java or C++.
Compile-time polymorphism is method overloading: several methods share a name but differ in parameter list, and the compiler picks one from the static types at the call site. Run-time polymorphism is method overriding: a subclass redefines a method of its superclass, and the actual implementation is chosen at execution time from the object’s dynamic type. In Java every non-static, non-final method is virtual by default, so overriding works automatically; in C++ you must mark the base method virtual, otherwise the call is resolved statically and you get the base version through a base pointer. The mechanism is a per-class virtual method table with a hidden pointer in each object, which is why a virtual call costs one extra indirection.
Q: Find the length of the longest substring without repeating characters.
Use a sliding window with two pointers and a hash map from character to its last seen index. Move the right pointer through the string; when you meet a character already inside the window, jump the left pointer to one past that character’s last index rather than moving it one step at a time. Update the answer as right - left + 1 at each step and update the map with the current index. Each character is visited at most twice, giving O(n) time, and space is O(min(n, k)) where k is the alphabet size - for lowercase ASCII that is effectively O(1). The brute-force alternative of checking every substring is O(n squared) or worse, so mentioning why the window never moves backwards is what earns the round.
Q: How does a HashMap work internally, and why must equals and hashCode agree?
A HashMap keeps an array of buckets; a key’s hashCode is spread and then mapped to a bucket index by masking against the array length, which is always a power of two. Collisions inside a bucket are chained in a linked list, and in modern Java a bucket that grows past a threshold of eight entries converts to a balanced tree, so worst-case lookup degrades to O(log n) rather than O(n). When the number of entries exceeds the load factor times capacity - 0.75 by default - the table doubles and every entry is rehashed. equals and hashCode must be consistent because get() first finds the bucket by hash and only then compares with equals: two objects that are equal but hash differently will land in different buckets and the map will silently lose the entry, which is also why mutating a key after insertion is a bug.
Q: What is normalization, and what are transactions and their isolation levels?
Normalization splits data to remove redundancy: 1NF demands atomic values, 2NF removes partial dependencies on part of a composite key, 3NF removes transitive dependencies between non-key columns, and BCNF tightens 3NF so every determinant is a candidate key. A transaction is a unit of work that is atomic, consistent, isolated, and durable - in SQL you wrap it in BEGIN, then COMMIT or ROLLBACK. Isolation levels trade safety for concurrency: Read Uncommitted allows dirty reads, Read Committed prevents them, Repeatable Read additionally prevents non-repeatable reads, and Serializable also prevents phantoms. Most production OLTP systems run at Read Committed and handle the remaining races with explicit locking or optimistic version columns.
Frequently asked questions about Barclays interviews
Section titled “Frequently asked questions about Barclays interviews”What is the Barclays interview process for freshers?
Barclays’ India campus technology process (BGSC) typically runs 3 stages: 1. An Online Assessment (~90 minutes) - MCQs on Computer Networks, OS, and DBMS, plus one SQL query and one medium-to-hard DSA problem. 2. A Technical Interview (~60 minutes) - roughly 25-30 minutes on your internship/academic projects, then OOP principles, programming fundamentals, and database concepts; some drives add a second technical round. 3. An HR round (~30 minutes) on resume, extracurriculars, situational questions, and which track (frontend/backend/database) you prefer.
What questions are asked in Barclays interviews?
The Online Assessment mixes Computer Networks, OS, and DBMS MCQs with a live SQL query and a medium-to-hard DSA problem. Technical interviews start with a genuine project deep-dive, then move into OOP concepts, Java/C++ fundamentals, and database questions. The HR round layers in RISES-values behavioral prompts (Respect, Integrity, Service, Excellence, Stewardship) using the STAR format, alongside standard motivation and fit questions.
How many rounds are there in the Barclays interview?
Most India campus drives run 3 stages after the resume shortlist: Online Assessment, one or two Technical Interviews, and an HR round. Barclays’ UK/international graduate-programme route instead runs numerical/verbal reasoning tests, a Situational Judgement Test, and a full-day Assessment Centre - a different pipeline from most India on-campus hiring.
What are Barclays’ RISES values, and how do they show up in interviews?
RISES stands for Respect, Integrity, Service, Excellence, and Stewardship - Barclays’ core values framework. HR and assessment-centre rounds map behavioral questions directly onto these values using the STAR method - for example, ‘describe a time you disagreed with a team but implemented the decision anyway’ (Respect/Service) or ‘how have you managed conflicting stakeholder priorities?’ (Stewardship). Knowing the five values by name and having one story mapped to each is genuinely useful prep.
How should I prepare for Barclays interviews?
Practice coding and aptitude for the online test, revise SQL (you’ll likely write queries live) plus Computer Networks/OS/DBMS fundamentals, and prepare a clear project narrative. For the HR round, read up on Barclays’ RISES values and have one STAR-format story ready for each - resonance with the values framework specifically is what interviewers are listening for.
Where is Barclays’ technology hiring based in India?
Through the Barclays Global Service Centre (BGSC), spread across four sites - Pune, Noida, Chennai, and Mumbai. Pune is Barclays’ largest site outside London and leads engineering, analytics, and shared services; Noida focuses on digital engineering and fintech solutions; Chennai handles back-office and platform support; and Mumbai covers investment banking and corporate finance technology.