GuidesUpdated 11 November 2026
Computer Science University Interview Questions and Formats
7 days free · cancel anytime
Computer science university admissions interviews at Oxford and Cambridge are academic conversations modeled on short tutorials, focusing on mathematical reasoning, logic, and algorithmic problem solving rather than programming syntax. At Oxford, each interview lasts around 30 minutes with two interviewers, while Cambridge applicants complete 1 or 2 interviews lasting 35 to 60 minutes in total during the main interview window of December 7 to December 18, 2026 (University of Oxford Department of Computer Science and University of Cambridge Undergraduate Study). Neither university requires prior coding experience, as tutors evaluate how you think out loud, test small base cases, and respond to hints.
Admissions Interviews vs Technical Job Interviews
AdmitMock is an AI mock interview app for Oxford, Cambridge, Imperial, UCL and US college alumni interviews: an AI interviewer asks the questions out loud, follows up, and shows you what to fix.
When you search for computer science university interview questions, two distinct types of conversations appear. One covers software engineering internship interviews for enrolled college undergraduates, focusing on language frameworks, system design, and database queries. The other covers undergraduate university admissions interviews for high school applicants applying to study computer science at institutions like Oxford and Cambridge.
Undergraduate admissions interviews do not test your knowledge of Python, Java, or C++. The University of Oxford Department of Computer Science explicitly notes that it accepts and encourages applications from candidates with little or no experience in programming computers. Similarly, the University of Cambridge Department of Computer Science and Technology confirms that interviewers will not expect you to know material outside your school curriculum.
Instead of writing code, you will spend your interview working through unfamiliar mathematical, logical, and combinatorial problems. Tutors want to see how you analyze an unfamiliar puzzle, how you communicate your thought process out loud, and how effectively you use an interviewer hint when your initial idea stalls.
Interview Format, Dates, and Structure for 2026-27
Both Oxford and Cambridge interview shortlisted computer science candidates in December after assessing UCAS applications and scores from the Test of Mathematics for University Admission (TMUA). Starting with the 2026 admissions round, Oxford replaced the Mathematics Admissions Test (MAT) with the TMUA, aligning with Cambridge so all applicants sit the TMUA following the October 15 UCAS deadline at 6pm UK time.
At Oxford, shortlisted candidates are invited to online interviews held on Microsoft Teams in early to mid-December 2026. Applicants generally have two interviews at their first-choice college and at least one interview at another college. Each Oxford computer science interview lasts around 30 minutes, usually with two interviewers. Under Corpus Christi College policy for Oxford online interviews, UK candidates are scheduled between 8:45am and 5:30pm, while overseas candidates interview between 7:30am and 10:00pm in their home time zone. Colleges aim to provide about one week of notice before subject interviews begin, and at least 24 hours of notice for the exact Microsoft Teams invitation. Shortlisted Oxford applicants for 2027 entry receive their admissions decisions on January 12, 2027.
At Cambridge, the main interview window runs from December 7 to December 18, 2026, with winter pool interviews taking place around mid to late January 2027 and mature applicant rounds from mid-March to early April. For the 2025 cycle, Cambridge Computer Science had 14 applications per place and accepted 128 students. Most Cambridge applicants have 1 or 2 interviews lasting a total of 35 minutes to an hour, though the exact setup depends on the college. By our calculation from the 2027 entry Cambridge College interview format table, 14 of 29 colleges interview all candidates online, 4 interview all candidates in person, and 11 use a location-based model where UK candidates interview in person and international candidates interview online.
Official Oxford and Cambridge Computer Science Interview Formats (2027 Entry)
| University / College | Number and Length of Interviews | Format and Platform | Required Admissions Test |
|---|---|---|---|
| Oxford (All Colleges) | Generally 2 at first-choice college and at least 1 at another college, around 30 minutes each | All online via Microsoft Teams (Tier 1 to Tier 3 whiteboard setup) | TMUA |
| Cambridge (Gonville and Caius) | Two 25-minute Computer Science interviews | UK and Ireland: choice of online or in person. Outside UK/Ireland: online | TMUA |
| Cambridge (Peterhouse) | One 25-minute Computer Science interview and one 25-minute Math interview | UK-based: in person. Outside UK: online | TMUA and CSAT |
| Cambridge (Churchill) | One 30 to 40 minute Computer Science interview | All in person | TMUA |
| Cambridge (Trinity) | 1 or 2 interviews (35 to 60 minutes total across Cambridge) | UK-based: in person. Outside UK: choice of online or in person | TMUA and CSAT |
Seven Official Sample Interview Problems
The University of Oxford Department of Computer Science publishes seven problems that have been used in past interviews, and the Cambridge Department of Computer Science and Technology recommends these exact problems for Cambridge preparation as well. They cover the primary areas assessed in academic interviews: combinatorics, search algorithms, game theory, parity invariants, and formal systems.
Tutors do not expect you to solve every stage of these problems immediately. Each question begins with accessible conditions before moving into deeper generalizations.
- Tidy boxes (Combinatorics and induction): You have 10 boxes that each hold exactly 10 wooden blocks, and 100 blocks in 10 different colors. There may not be the same number of blocks in each color. Show that it is possible to pack the blocks so that each box contains at most two different colors.
- Searching for the maximum (Search algorithms and complexity): A real-valued function f(x) defined for x from 0 to 1 has a single maximum at x = m. For u less than v up to m, f(u) is less than f(v), and for u less than v from m to 1, f(u) is greater than f(v). You know nothing else about f, but you can ask for the value of f(x) at any points you choose. How would you find the approximate value of m, and how accurately could you find it using only 10 evaluations?
- Death by chocolate (Game theory and symmetry): You and your enemy are locked in a room with a rectangular bar of chocolate divided into squares. One square is painted with deadly green poison. You take turns breaking off one or more squares along a straight line and eating them. Whoever is left with the green square must eat it. You can inspect the bar and decide whether to go first or second. Describe your strategy.
- Monkey beans (Parity invariants): An urn holds 23 white beans and 34 black beans. A monkey takes out two beans at a time. If they are the same color, he puts a black bean into the urn from a large heap next to him. If they are different colors, he puts in a white bean. He repeats this until one bean is left. What color is the final bean?
- Lily-pad lunacy (Recurrence relations): Eleven lily pads are numbered 0 to 10. A frog starts on pad 0 and wants to reach pad 10 by jumping forward 1 or 2 pads at a time. Order matters, so jumping 221212 is different from 221122. How many different ways are there to get from pad 0 to pad 10?
- Missing numbers (Algorithmic efficiency and space complexity): You are given a list of slightly less than 1,000,000 numbers, all different, each between 0 and 999,999 inclusive. How could you find, in a reasonable time, a number between 0 and 999,999 that is not on the list?
- Scribble (Formal rule systems and state invariants): The game begins with the word PQ on a board using tiles P, Q, and R. Rule 1 allows changing Px to Pxx for any shorter word x. Rule 2 replaces QQQ with R anywhere in the word. Rule 3 deletes RR anywhere in the word. Determine whether you can make QPR, PQQ, PQR, and PR, and explain how to decide if any given word can be made.
How a Real Interview Unfolds: The Tidy Boxes Walkthrough
Solving a problem on paper is completely different from talking through it with university tutors. Oxford publishes an official sample interview dialogue for the Tidy Boxes problem that both Oxford and Cambridge highlight as a model of a successful interview.
The conversation illustrates how an interview evolves through five distinct stages, transitioning from an initial roadblock to small base cases, an inductive reduction, an algebraic bound, and a counterexample.
- Stage 1, Communicating the roadblock: The candidate begins by explaining their instinct to place each color into its own box. This leaves gaps in boxes holding colors with fewer than 10 blocks and creates surplus blocks for larger colors, but the candidate admits they do not see how to match surpluses to gaps. Stating where your idea breaks down is far more helpful than remaining silent.
- Stage 2, Testing base cases (n = 1 and n = 2): The interviewer simplifies the scenario to 1 box with 10 blocks of 1 color, followed by 2 boxes with 20 blocks in 2 colors. Both are trivial, since 2 boxes can accommodate 20 blocks of 2 colors in any distribution without violating the two-color rule.
- Stage 3, Inductive reduction using averages: Returning to 10 boxes and 100 blocks across 10 colors, the interviewer prompts the candidate to fill just the first box. Because the average number of blocks per color is 10, at least one color must have 10 or fewer blocks, and at least one must have 10 or more blocks. Putting all of the smaller color into the box leaves a gap of at most 9 blocks, which can be filled using part of the larger color. Setting that completed box aside leaves 9 boxes, 90 blocks, and 9 colors, reducing the problem down to 1 box.
- Stage 4, Generalizing with algebra (11 colors): The interviewer increases the difficulty to 10 boxes and 11 colors. When the candidate questions whether a gap of 9 can always be filled when fewer boxes remain, the tutor suggests using algebra. With n boxes and n plus 1 colors left, if the smallest color has x blocks where x is at most 10, the gap is 10 minus x. The remaining 10n minus x blocks are divided among n colors, giving an average of (10n minus x) / n, which equals 10 minus x/n. Because n is at least 1, 10 minus x/n is always at least as large as 10 minus x.
- Stage 5, Proving limits with a counterexample (12 colors): For 10 boxes and 12 colors, the candidate recognizes that eliminating one color per box leaves 1 box and 3 colors at the end. To prove the problem cannot always be solved, the candidate and tutor construct a concrete counterexample: 11 colors with 1 block each and 1 color with 89 blocks. By the pigeonhole principle, placing 11 single blocks into 10 boxes forces at least two single blocks into one box, leaving 8 empty slots that must be filled by the large color, creating a box with 3 colors.
Worked Solutions to the Other Sample Problems
Tutors expect you to use standard problem-solving techniques when answering interview prompts: testing minimal inputs, searching for invariants, and analyzing algorithmic complexity. Here is how our team applies official Oxford hints to the remaining sample problems.
- Monkey beans: Example written by our team. The applicant is invented. Test a minimal case such as 1 white and 1 black bean. In every move, the monkey removes two beans and replaces them with one, altering the white bean total by either 0 or minus 2. Because the starting count of 23 white beans is odd, its parity never changes, meaning the final bean must be white.
- Lily-pad lunacy: Start at small values instead of pad 10. There is 1 way to reach pad 1 and 2 ways to reach pad 2. To reach pad n, the frog makes a final leap of 1 from pad n minus 1, or a leap of 2 from pad n minus 2. This creates the recurrence relation F(n) = F(n-1) + F(n-2). Stepping through the Fibonacci sequence yields 1, 2, 3, 5, 8, 13, 21, 34, 55, and 89 ways to reach pad 10.
- Death by chocolate: Examine a 1-row bar first. If the poison is at one end, the first player breaks off all safe squares, leaving the second player to eat the poison. For a square n by n grid with the poison square in a corner, the second player can maintain an L-shape with symmetric arms, copying every move made by the first player on the opposite arm to secure a win.
- Missing numbers: If exactly one number is missing from the range 0 to n, sum all integers from 0 to n using n(n+1)/2 and subtract the input total. This achieves O(n) time complexity and O(1) extra space. If multiple numbers are absent, a bit array uses O(n) time and O(n) bits. Alternatively, partitioning the list by counting how many values fall below 500,000 enables a binary search over the range in 20 passes using O(1) additional memory.
- Searching for the maximum: Because you cannot evaluate derivatives, a single midpoint check cannot show which direction slopes upward. You must evaluate two points u and v where u is less than v. If f(u) is less than f(v), the maximum cannot lie in [0, u], shrinking the interval to [u, 1]. A ternary or golden-section search narrows the range by a constant factor on each step, achieving logarithmic query complexity across the 10 evaluations.
- Scribble: The letter P only appears at the start of a string, so QPR can never be generated. Next, represent the rules using modular arithmetic. Rule 2 replaces QQQ with R, and Rule 3 deletes RR, meaning R equals 3 Qs and 6 Qs cancel out. Starting with PQ yields 1 mod 6. Rule 1 doubles the string after P, which doubles the value modulo 6. Powers of 2 modulo 6 produce only 2 and 4, meaning any word whose Q-equivalent count is divisible by 3 cannot be formed.
Technical Personal Statement Defense
Oxford and Cambridge interviewers frequently begin with questions about projects, independent reading, or tools mentioned in your UCAS personal statement. The Cambridge Department of Computer Science and Technology explicitly advises candidates to review everything in their statement prior to the interview, particularly coding projects and algorithms.
Tutors will not ask for memorized buzzwords. They will ask you to justify your architectural choices, discuss how your code handled errors, and evaluate how your data structures scale with larger input sizes.
- Review your codebases: Read through your project code so you can describe your functions, control flow, and logic clearly from memory.
- Justify data structures: Be prepared to explain why you chose an array, hash table, tree, or linked list, and the computational trade-offs involved.
- Discuss bugs openly: Tutors value candidates who can explain where their project broke, how they isolated the issue, and what changes they would make today.
- Connect reading to principles: If you cited a text on machine learning or algorithms, focus on explaining the mathematical foundation rather than summarizing chapter titles.
Online Interview Setup and Whiteboard Guidance
Undergraduate admissions interviews at Oxford take place online via Microsoft Teams, categorized into three technology tiers. Tier 2 uses an interactive virtual whiteboard alongside Teams, and Tier 3 adds handwriting capture using a touchscreen and stylus. Cambridge colleges also ask science and mathematics candidates to share their workings on screen.
Effective whiteboard communication ensures that your visual workings support your verbal explanations without causing delays.
- Narrate your calculations: Avoid writing silently. Speak through your mathematical workings and logic as you write so the interviewers can track your reasoning.
- Organize your layout: Keep base cases on the left side, inductive steps or algorithmic loops in the center, and generalized results on the right.
- Sketch clean diagrams: When explaining trees, graphs, or state transitions, draw clear nodes and directed edges rather than crowded sketches.
- Keep physical paper on hand: Have a dark pen and notepad nearby. If your stylus disconnects or the software stutters, you can hold handwritten workings up to your camera.
A 14-Day Computer Science Interview Preparation Plan
This two-week preparation plan was developed by our team to help you organize your practice using official University of Oxford and University of Cambridge materials.
- Days 1 to 3, Verbal problem solving: Work through Oxford's seven official sample interview problems. Practice replacing large values with 0, 1, and 2 out loud with a friend or mock interview tool.
- Days 4 to 6, Discrete mathematics and induction: Work through past paper problems from the TMUA and MAT covering logic, sequences, and proof by induction. Review official MAT Computer Science livestreams for model solutions.
- Days 7 to 9, Algorithmic thinking: Solve problem-solving prompts from Project Euler and British Informatics Olympiad starter tasks. Analyze each method for Big-O time and space complexity.
- Days 10 to 11, Personal statement audit: Review your UCAS personal statement line by line. Draft explanations for every project, library, and algorithm you cited, detailing why you chose them over alternatives.
- Days 12 to 13, Mock interview practice: Complete a full 30-minute timed mock interview using video conferencing and a shared virtual whiteboard, practicing writing with a stylus while explaining your steps out loud.
- Day 14, Final setup check: Test your webcam, microphone, internet connection, and stylus battery. Review the interviewees' code of conduct and confirm your interview time zone.
Practice example
An urn contains 23 white beans and 34 black beans. A monkey takes out two beans at random. If they are the same color, he puts a black bean into the urn from a large pile next to him. If they are different colors, he puts a white bean into the urn. He repeats this until only one bean is left. What color is the final bean, and how can you be sure?
Example written by our team. The applicant is invented. Let me start by checking what happens in each step, and test a smaller number of beans first, such as 1 white and 2 black beans. In every step, the monkey removes two beans and puts one back, so the total number of beans in the urn decreases by 1 each time. Now let me look at how the count of each color changes across the three possible pairs the monkey can draw. First, if he draws two black beans, he puts one black bean back, so black decreases by 1 and white changes by 0. Second, if he draws two white beans, he puts one black bean back, so white decreases by 2 and black increases by 1. Third, if he draws one white and one black bean, he puts one white bean back, so black decreases by 1 and white stays the same. In all three cases, the number of white beans changes by either 0 or minus 2. That means the parity of the white beans never changes. Because we started with 23 white beans, which is odd, there must always be an odd number of white beans in the urn. When only 1 bean remains, the final bean must be white.
The candidate follows Oxford's official guidance by thinking out loud, proposing a small test case first, and systematically listing the three possible transitions before jumping to a conclusion. By identifying that the number of white beans changes only by 0 or minus 2, the candidate turns a seemingly random process into a clean parity invariant proof.
Frequently asked questions
Do I need to know how to code for an Oxford or Cambridge computer science interview?
No. The University of Oxford Department of Computer Science states that it accepts and encourages applications from candidates with little or no programming experience, and Cambridge interviewers do not test material outside your school syllabus. Interviews focus on mathematical reasoning, logic, and problem-solving method rather than coding syntax.
What happens if I get stuck on a problem during my computer science interview?
Getting stuck is completely normal, as interviewers do not expect you to complete every problem presented. Tutors provide hints when you stall or take an unproductive direction to observe how you process feedback. Explaining your thought process out loud allows interviewers to provide timely guidance.
How long are Oxford and Cambridge computer science interviews?
At Oxford, each interview lasts around 30 minutes with two interviewers, and candidates generally have two interviews at their first-choice college plus at least one at another college. At Cambridge, most applicants complete 1 or 2 interviews lasting 35 to 60 minutes in total, such as two 25-minute interviews at Gonville and Caius or one 30 to 40 minute interview at Churchill.
Are computer science university interviews online or in person for 2027 entry?
All Oxford undergraduate interviews take place online via Microsoft Teams in December 2026. At Cambridge, format varies by college: 14 colleges interview all applicants online, 4 interview all applicants in person, and 11 use a location-based model where UK candidates interview in person and overseas applicants interview online.
What admissions tests are required before a computer science interview?
All computer science applicants to Oxford and Cambridge must register for and take the Test of Mathematics for University Admission (TMUA), which replaced the MAT at Oxford starting with the 2026 admissions cycle. Cambridge applicants to Peterhouse or Trinity College must also sit the Computer Sciences Aptitude Test (CSAT).
Related guides
- Electrical Engineering University Interview Questions and Preparation Guide
- Film Production University Interview Questions: Portfolio, Crew, and Course Fit
- Fine Art University Interview Questions: Format, Scoring, and Practice
- Graphic design uni interview questions: what tutors ask and how to prepare
- How Long Should an Interview Last?
- All admissions interview guides
Sources
Checked 11 November 2026.
- University of Oxford Department of Computer Science: Sample Interview Problems
- University of Oxford Department of Computer Science: Sample Interview Dialogue
- University of Oxford Department of Computer Science: How to Apply
- University of Oxford Undergraduate Admissions: Interviews Guide
- Corpus Christi College Oxford: Undergraduate Interview Policy
- University of Cambridge Department of Computer Science and Technology: Interviews
- University of Cambridge Undergraduate Study: Computer Science BA (Hons) and MEng
- University of Cambridge Undergraduate Study: What to Expect at Your Cambridge Interview
- cs.ox.ac.uk
AdmitMock is not affiliated with Oxford, Cambridge or any other university.

