Skip to content

Deutsche Bank Interview Questions and Answers (2026)

Deutsche Bank’s technology graduate loop layers a HackerRank coding OA on top of standard SHL assessments, then closes with a HireVue screen and a Superday of live technical interviews about a week later.

Deutsche Bank interview process at a glance

Section titled “Deutsche Bank interview process at a glance”
Round Duration What they test
SHL online assessments ~55-75 min total Numerical, verbal, inductive/logical reasoning, situational judgement
HackerRank coding OA (technology track) ~90 min 10 MCQs (DS, C, OS, graph theory) + 2 coding problems
HireVue Video Interview 3-5 questions Recorded behavioral answers, ~3 min each
Superday 3 back-to-back interviews DSA, CV-based technical questions, domain fit, behavioral

A standard graduate-scheme screen shared across Deutsche Bank’s business lines: numerical reasoning (20 questions, 20 min), verbal reasoning (20 questions, 20 min), inductive/logical reasoning (15 min), and a situational judgement test on workplace scenarios.

Common questions

  • Data-interpretation and numerical-reasoning problems under time pressure
  • Verbal reasoning - passage-based true/false/cannot-say questions
  • Inductive/pattern-based logical reasoning
  • Situational judgement scenarios on teamwork and prioritisation

Technology-track applicants sit an additional coding OA: 10 technical MCQs spanning data structures, C, operating systems, and graph theory, plus 2 programming problems, in roughly 90 minutes.

Common questions

  • Data-structure and OS MCQs (scheduling, memory management, complexity)
  • Graph-theory MCQs (traversals, shortest path concepts)
  • Valid parentheses / expression checks (stack-based)
  • Find duplicates in a stream (hashing)

A recorded, one-way interview with 3-5 pre-set questions - 30 seconds to prepare and about 3 minutes to record each answer.

Common questions

  • Why Deutsche Bank, and why technology within a bank?
  • Tell me about a time you handled a tight deadline or a critical bug
  • Describe a teamwork or leadership experience
  • What do you know about Deutsche Bank’s business and technology focus?

Roughly a week after HireVue, 3 back-to-back live interviews with the team - the most CV-driven stage of the process, where interviewers build follow-up questions directly from the projects and technologies on your resume.

Common questions

  • SQL + coding hybrid - nth-highest-salary style queries using window functions
  • Curveball technical questions based on a specific project on your CV
  • Explain settlement systems or a Java concept in plain language
  • Detailed project discussion - architecture, hardest bug, what you’d change

Full round-by-round narratives are on the Deutsche Bank interview experience page.

Unlike the automated SHL and HackerRank stages earlier in the funnel, Deutsche Bank’s Superday interviewers consistently build technical follow-ups straight from the exact projects and technologies listed on your CV rather than pulling from a generic question bank. Candidates report being asked to defend specific implementation choices - why a particular data structure, why that stack, what would break at scale - on projects they listed, sometimes months after writing the resume. Re-reading your own CV line by line and rehearsing a defensible answer for every technology you claim is one of the highest-leverage things to do before the Superday.

Common technical interview questions and answers

Section titled “Common technical interview questions and answers”
Q: How do you check whether an expression has balanced parentheses?

Scan the string once with a stack. Push every opening bracket; on every closing bracket, pop the top and check it is the matching opener - if the stack is empty or the pair mismatches, the expression is invalid. After the scan the expression is balanced only if the stack is empty, which catches unclosed openers. Time complexity is O(n) and space is O(n) in the worst case, where every character is an opening bracket. Deutsche Bank’s HackerRank OA asks this as a valid-parentheses or expression-check problem, and interviewers often follow up by asking you to extend it to three bracket types or to report the index of the first mismatch.

Q: How do you find duplicates in a stream of numbers?

Keep a hash set of values already seen. For each incoming element, check membership in O(1) average time: if it is present, report a duplicate, otherwise insert it. Over n elements this is O(n) time and O(k) space for k distinct values. If the values are bounded - say integers from 1 to n - you can drop the hash set and use a bit array or the in-place index-negation trick for O(1) extra space. If the stream is unbounded and approximate answers are acceptable, a Bloom filter gives constant memory with a tunable false-positive rate but never false negatives.

Q: Write a SQL query to find the third-highest salary.

Use a window function: SELECT DISTINCT salary FROM (SELECT salary, DENSE_RANK() OVER (ORDER BY salary DESC) AS rnk FROM employees) t WHERE rnk = 3; DENSE_RANK is the right choice because it gives tied salaries the same rank and does not skip the next number, so the third-highest distinct salary really is rank 3. Using RANK instead would skip ranks after a tie and could return no row, and ROW_NUMBER would treat duplicate salaries as distinct values. The older correlated-subquery or LIMIT 1 OFFSET 2 forms also work, but window functions are what Deutsche Bank interviewers usually want to see.

Q: What is the difference between RANK, DENSE_RANK and ROW_NUMBER?

All three are window functions that number rows within a partition according to an ORDER BY. ROW_NUMBER assigns a unique sequential number even to ties, so two equal salaries get 1 and 2 arbitrarily. RANK gives ties the same number but then skips values - two rows tied at 1 are followed by rank 3. DENSE_RANK gives ties the same number and does not skip, so the same case is followed by rank 2. Pick ROW_NUMBER for deduplication or pagination, RANK when gaps are meaningful in a leaderboard, and DENSE_RANK for nth-distinct-value queries.

Q: What is trade settlement, and what does T+1 mean?

Settlement is the post-trade stage where the securities actually move to the buyer and the cash moves to the seller, completing the obligation created when the trade was executed. T+1 means settlement finalises one business day after the trade date - Indian equities moved to T+1 in 2023 and US markets followed in May 2024, replacing the older T+2 cycle. Shorter cycles reduce counterparty credit risk and the margin clearing houses must hold, but they compress the window for confirmation, allocation and funding, which is exactly why banks invest heavily in settlement technology. A clearing house acts as central counterparty, netting obligations so far fewer gross transfers are required.

Q: What is the difference between a process and a thread, and what happens on a context switch?

A process has its own virtual address space, file descriptors and page tables; threads live inside a process and share its heap, globals and open files while keeping their own stack, registers and program counter. That sharing makes thread creation and communication far cheaper than inter-process communication, but it also means data races are possible without synchronisation. A context switch saves the current execution context - registers, program counter, stack pointer - into the process or thread control block and restores the next one. Switching between processes is more expensive than between threads because it also swaps the address space, which flushes or invalidates TLB entries and hurts cache locality.

Q: When would you use BFS instead of Dijkstra’s algorithm?

Use plain BFS when every edge has the same weight, because BFS explores vertices in nondecreasing distance order and finds shortest paths in O(V + E) time with a simple queue. Dijkstra’s algorithm handles arbitrary non-negative edge weights using a priority queue, costing O((V + E) log V) with a binary heap - strictly more work than BFS on an unweighted graph. For graphs whose edges are only 0 or 1, a deque-based 0-1 BFS gets Dijkstra’s result in O(V + E). Neither handles negative edge weights: for those you need Bellman-Ford at O(V * E).

Q: What is the difference between HashMap, Hashtable and ConcurrentHashMap in Java?

HashMap is unsynchronised, permits one null key and multiple null values, and gives O(1) average get and put; since Java 8 a bucket that grows past eight entries converts from a linked list to a balanced tree, so the worst case is O(log n) rather than O(n). Hashtable is the legacy class that synchronises every method on the whole object, so it serialises all access and is effectively obsolete. ConcurrentHashMap allows concurrent reads without locking and locks only individual bins on write, so throughput scales with cores; it forbids null keys and values so that a null return unambiguously means absent. Use ConcurrentHashMap for shared mutable maps and plain HashMap for thread-confined ones.

Frequently asked questions about Deutsche Bank interviews

Section titled “Frequently asked questions about Deutsche Bank interviews”
What is the Deutsche Bank interview process for freshers?

Deutsche Bank’s Technology, Data and Innovation (TDI) graduate loop typically runs: 1. SHL online assessments - numerical reasoning (20 questions, 20 min), verbal reasoning (20 questions, 20 min), inductive/logical reasoning (15 min), and a situational judgement test. 2. A HackerRank coding OA for technology roles - 10 technical MCQs (data structures, C, OS, graph theory) plus 2 programming problems, about 90 minutes total. 3. A HireVue recorded video interview - 3-5 questions, 30 seconds to prepare and about 3 minutes to record each. 4. A Superday roughly a week later - 3 back-to-back in-person (or virtual) interviews with the team, covering technical depth, resume projects, and behavioral fit. Total duration: roughly 3-4 weeks from application to offer.

What questions are asked in Deutsche Bank interviews?

The HackerRank OA mixes MCQs on data structures, C, operating systems, and graph theory with 2 coding problems. Technical interviews commonly ask stack-based problems (valid parentheses/expression checks), hashing problems (find duplicates in a stream), and SQL-plus-coding hybrids (nth-highest-salary style queries, window functions), plus curveball technical questions built directly from projects listed on your CV. Expect domain conversation around settlement systems, Java, and SQL explained in plain language. Behavioural rounds check ownership, teamwork, and why Deutsche Bank.

How many rounds are there in the Deutsche Bank interview?

Deutsche Bank’s TDI stream typically runs 4-5 stages: SHL online assessments, a HackerRank coding OA (technology track only), a HireVue video interview, and a Superday with 3 back-to-back interviews about a week later. Some drives combine or skip a stage - check that cycle’s college placement communication or offer email for the exact sequence.

What is the Deutsche Bank Superday and how is it different from earlier rounds?

The Superday is Deutsche Bank’s final-round format, typically scheduled about a week after the HireVue stage: 3 back-to-back interviews with the team in one sitting, mixing technical depth, resume-project cross-questioning, and behavioural fit. Unlike the earlier automated OA and HireVue stages, every Superday interview is a live conversation with an actual Deutsche Bank technologist, so being able to defend your project decisions and explain trade-offs matters more here than raw problem count.

How should I prepare for Deutsche Bank interviews?

Practise timed DSA (stacks, hashing) and revise OS, data structures, and graph-theory fundamentals for the HackerRank MCQ section specifically. Revise SQL window functions and joins, prepare one crisp project narrative you can defend under curveball questions about implementation choices, and read up on settlement systems and Java so domain conversation doesn’t feel like buzzwords. Rehearse the HireVue format - short, structured answers recorded in a single take - and use STAR for behavioural answers.

Is Deutsche Bank’s process different for the Technology, Data and Innovation (TDI) stream vs other graduate roles?

Yes. The TDI stream adds a HackerRank coding OA (10 MCQs plus 2 programming problems) on top of the standard SHL online assessments that all Deutsche Bank graduate applicants take, and its interviews lean heavily on curveball technical questions based on your CV projects. Non-technology graduate streams skip the coding OA entirely and focus more on case-style reasoning and finance fundamentals.

Looking for placement papers, OA practice, or coding questions?

Section titled “Looking for placement papers, OA practice, or coding questions?”