American Express runs two visibly different loops depending on where you apply. Its large India technology and analytics organisation hires mostly on campus: a timed Codility (or Unstop) online assessment, then two to three interviews that mix data structures, DBMS/SQL, OOP and core CS with resume and project drilling, puzzles and guesstimates, closing with a managerial or HR conversation. Its US, UK and Phoenix corporate, finance and strategy pipeline is behavioural-first: a HireVue video or written screen, then short 30-45 minute one-on-ones, an assessment centre or group case in the UK, and repeated business questions about how Amex actually makes money. Across both tracks, candidates who cannot explain the card revenue model or answer 'why Amex' get filtered out, and interviewers consistently reward a visible, structured thought process over a perfect answer.
These 13 writeups cover software engineering, data & science, analytics and other roles, India campus + US/UK corporate. All of them were posted under a pseudonym (Reddit, GeeksforGeeks or forum handles), so we could not verify the authors’ identities.
Typical rounds
3.21-5 range
Outcomes shared
5/4offer / not
Most common round
Technical
Sources span
2015 - 2025
Most frequently reported · Behavioral(10), Data structures & algorithms(9), Domain knowledge(9), Past projects(8), Resume deep-dive(7)
American Express’s official process
Amex publishes region-by-region student hiring steps. In the US the sophomore finance internship starts with recorded video responses to behavioural questions and then two 30-minute HireVue interviews, with an added technical assessment for technology roles. Canada runs two hiring-leader interviews followed by a recruiter phone screen. UK and European programmes use a pre-recorded video followed by an assessment centre of competency interviews, case studies and group discussions, with technology apprentices taking an online coding assessment before a virtual discovery day. India is described as university-recruitment led, and Singapore uses an on-site group assessment task plus a hiring-leader interview. Amex also publishes AI guidelines stating candidates may use AI to prepare but not during a live interview unless the interviewer authorises it.
Three coding problems on Codility in 90 minutes. Most test cases are hidden, so candidates cannot tell whether a solution is fully correct before submitting.
Covered · Data structures & algorithms, Arrays & strings
2
Technical interview 1Technical
Introduction and a walk through the resume, then a long stretch of C/C++ and object-oriented fundamentals covering polymorphism, constructors and destructors, inheritance and overloading versus overriding, plus a motivation question about Amex.
Covered · OOP, Past projects, Resume deep-dive
3
Technical interview 2Technical
Deeper systems questions on keys, ACID properties and database locking, exception handling and access specifiers, tree types, a small swap-without-temp program, some Python and tricky Java output questions, and networking basics.
Covered · DBMS, SQL, Networking, Trees, Debugging
4
HR interviewBehavioral
A pure HR conversation lasting over half an hour, opening with an explanation of the company's business and moving into situational questions about coping with failure, handling disputes and a decision the candidate later regretted.
A 90-minute Codility set of three problems, one on graphs, one using a priority queue and one easy warm-up, with hidden test cases and speed of submission reportedly taken into account.
Covered · Graphs, Data structures & algorithms
2
Technical interviewTechnical
An hour-long Webex interview where the candidate shared their screen and coded array reversal and linked-list creation and traversal live, explaining approach and complexity, after discussing their web development projects.
Covered · Arrays & strings, Data structures & algorithms, Past projects
3
Technical + HR interviewHiring manager
A roughly 50-minute mixed round where the interviewer asked for a live walkthrough of a project and then posed open scenario questions, including how the candidate would keep a year-long project profitable.
Covered · Past projects, Domain knowledge, Behavioral
Three easy-to-medium problems in 90 minutes drawing on hash maps, graphs, sets and string handling.
Covered · Graphs, Arrays & strings, Data structures & algorithms
2
Technical interview 1Technical
Language fundamentals in C++ and Java plus object-oriented concepts, closing with an implementation of a stack or queue backed by arrays.
Covered · OOP, Data structures & algorithms
3
Technical interview 2Low-level design
A schema design exercise for a school, with the interviewer layering on new situations as the design evolved, then cycle detection in a linked list, project questions, BFS and how hash maps store data and resolve collisions.
Covered · Low-level design, DBMS, Graphs, Past projects
4
HR interviewBehavioral
A reflection on how the technical rounds went, followed by situational questions drawn from the candidate's internship and projects and a discussion of future plans.
A nominal resume screen at an IIT campus drive that in practice let every applicant through to the online test.
Covered · Resume deep-dive
2
Online testOnline assessment
An Unstop-hosted multiple-choice test split into numerical ability, logical reasoning and data interpretation, and coding, at easy to medium difficulty.
Covered · Aptitude, Data structures & algorithms
3
Technical + HR interview 1Technical
Project discussion where honesty about a limited personal contribution was received well, then reading and drawing inferences from two SQL tables, a market-sizing guesstimate on annual snack packet sales, and questions on what the candidate knew about Amex.
Covered · SQL, Market sizing, Past projects, Domain knowledge
4
Technical + HR interview 2Final
The interviewer asked for two things that mattered most on the resume, then a guesstimate on the mass of the International Space Station and a weighing puzzle identifying the odd box of balls in the fewest weighings.
A compulsory 30-minute aptitude section plus optional machine learning and business case sections, followed by a separate psychometric section that repeats similar questions with different framing.
Covered · Aptitude, Case study
3
Technical + HR interview 1Technical
Machine learning fundamentals including clustering and dimensionality reduction, applied to a scenario about compressing a high-dimensional card-transaction feature vector without losing accuracy.
Covered · Domain knowledge, Past projects
4
Technical + HR interview 2Technical
A debrief on the hardest question from the previous round, then a deep dive into a CV project's algorithms and approach, plus a lateral-thinking task about generating random numbers in a real-world setting without code.
Covered · Past projects, Puzzles
5
Technical + HR interview 3Final
More project discussion alongside classic logic puzzles, and general HR questions extending into how Amex earns money, its business model and its different teams.
Covered · Puzzles, Past projects, Domain knowledge
Three virtual Codility problems, one easy and two medium, all solved before advancing.
Covered · Data structures & algorithms
2
Technical interviewTechnical
An in-person round that revisited an assessment problem and added classic DP and subarray questions, a merge sort trace including the integer-overflow flaw in the midpoint formula, OOP pillars in Java, process/thread/context-switch and critical-section questions, SQL versus NoSQL with ACID and CAP, a concurrent-purchase consistency scenario, indexing, and MERN front-to-back communication.
A conversational round about background and college life plus a debrief of the previous interview, then resume follow-ups on serverless and prompt engineering, and two written SQL queries for the second-highest and the tenth-to-fifteenth highest salaries.
Separate tests per track. The analytics track paired two short coding problems with sixty sectionally-timed MCQs across aptitude, English, C++, Java, SQL and DSA, while the technology track ran three harder coding problems, and completion speed fed the shortlist.
Covered · Aptitude, SQL, Data structures & algorithms, Dynamic programming
2
Technical + HR interviewPanel
A near-hour two-interviewer round that dug into a deep learning internship, added in-depth ML questions, a modified probability puzzle, SQL on deduplication and joins, OOP concepts, several questions on Amex itself, an open design question on expanding into a new geography, and how card issuers make money.
Covered · Domain knowledge, SQL, OOP, Puzzles, Case study
3
Technical + HR + behavioralFinal
A roughly 50-minute round with a VP covering every line of the resume, a heavy load of behavioural questions, and open discussion of machine learning, AI and voice assistants.
Around twenty students were shortlisted purely on resume strength, CGPA and projects, with no written test.
Covered · Resume deep-dive
2
Technical interview 1Technical
A light project walkthrough followed by three classic logic puzzles and then SQL questions on joins with a few queries to write out.
Covered · Puzzles, SQL, Past projects
3
Technical interview 2Technical
Project motivation and difficulties, then business questions on Amex's revenue model, familiarity with Excel and more SQL including altering a table. The candidate was rejected after failing to explain how Amex makes money.
Three Codility problems ranging from medium to hard, with the shortlist announced about a month later.
Covered · Data structures & algorithms
2
Technical interviewPanel
A 90-minute two-interviewer panel focused on the candidate's backend internship, npm versioning and React hooks, a duplicates-in-an-array problem discussed with two different approaches, joins and indexing, a hats-and-deduction logic puzzle, and motivation questions about Amex.
Covered · Past projects, Arrays & strings, DBMS, Puzzles, API design
3
Managerial + HRHiring manager
An engineering manager mixed technical and fit questions: how a request travels from front end to back end, authentication and JWT, REST and HTTP, and then strengths, weaknesses and how the candidate resolves group disputes.
Two coding problems in 30 minutes, one a sort-and-binary-search counting task and one lowest common ancestor in a binary tree, followed by a 50-question section in 40 minutes spanning aptitude, logical reasoning, data structures, JavaScript and C/C++.
Covered · Data structures & algorithms, Trees, Aptitude
2
Technical interviewTechnical
A friendly round starting with projects and internships, then self-join style SQL over an employee-manager table, C/C++ basics such as dangling pointers and static functions, a sieve program and a few standard puzzles.
Covered · SQL, Puzzles, Past projects
3
Semi-technical caseCase study
A business case about how Amex could partner with a hotel chain around a major sporting event, plus one data-structure question drawn from the resume.
Covered · Case study, Domain knowledge, Resume deep-dive
4
HRBehavioral
General fit questions on motivation for Amex, strengths and weaknesses, plans for further study and which team the candidate would pick.
A HireVue stage that unusually asked for written rather than recorded answers.
Covered · Behavioral
2
Four one-on-one interviewsCase study
Four 45-minute interviews each split roughly into fifteen minutes of behavioural questions and thirty minutes of casework, with early cases generic and later ones tailored to Amex and enterprise strategy.
Covered · Case study, Behavioral, Domain knowledge
A behavioural opener covering self-introduction, motivation for Amex and teamwork, then Amex-specific questions on how the company makes money and a hypothetical advising scenario about launching a card aimed at college students.
Covered · Behavioral, STAR stories, Domain knowledge, Case study
What is the interview process like at American Express?
American Express runs two visibly different loops depending on where you apply. Its large India technology and analytics organisation hires mostly on campus: a timed Codility (or Unstop) online assessment, then two to three interviews that mix data structures, DBMS/SQL, OOP and core CS with resume and project drilling, puzzles and guesstimates, closing with a managerial or HR conversation. Its US, UK and Phoenix corporate, finance and strategy pipeline is behavioural-first: a HireVue video or written screen, then short 30-45 minute one-on-ones, an assessment centre or group case in the UK, and repeated business questions about how Amex actually makes money. Across both tracks, candidates who cannot explain the card revenue model or answer 'why Amex' get filtered out, and interviewers consistently reward a visible, structured thought process over a perfect answer.
What topics does American Express test in interviews?
Commonly reported topics include Data structures & algorithms, SQL, DBMS, OOP, Puzzles, Aptitude.
These guides summarize public, first-hand interview experiences shared by candidates. They describe the shape of each loop and the topics that came up, not a leaked question bank, and every experience links back to its original source. Processes change often; treat this as directional prep, not a script.