Your American Express Prep Plan

You have seen how the software loop at American Express is built and judged. Now turn that into a plan you can run. One thing shapes the order more than anything else. Amex doesn't interview like a pure algorithm shop. The coding bar is real, but it tops out around LeetCode medium. Candidates get separated on Java and Spring depth, practical build-and-run coding, the payments domain, and whether they understand the business. So the plan below doesn't pour every hour into hard graph problems. It front-loads the coding that earns the next round. Its widest block goes to the Java internals and Spring ecosystem that a LeetCode-only prep skips. It closes on the design and behavioral work, which this loop treats as a genuine gate. The plan follows the India loop first. US pipelines tend to be lighter, so confirm yours with your recruiter.

The American Express angle

Before you drill anything, place your req, because the entry point splits here. Campus or off-campus candidates, interns and freshers open with a timed online assessment they must clear before meeting anyone. Experienced or lateral hires usually skip the assessment and start on a technical screen. A recruiter often briefs them on the topics to expect first. Read the interview process and how leveling works so both are settled before you plan your weeks. Note the ladder gotcha while you are there. Amex numbers its engineering titles in reverse, so Engineer III is the entry rung and Engineer I is the senior one. Knowing that keeps you from misreading a req or a recruiter note.

• The screen: LeetCode-medium DSA, coded in a plain editor. Arrays and strings, sorting such as the one-pass Dutch-flag partition, matrix rotation, stack and queue implementation, and backtracking such as a Premium question. Mostly easy-to-medium with the occasional harder ceiling. Often coded in a plain editor or a code-it-and-run-it window instead of a polished judge, so fluent Java without a full IDE matters as much as the algorithm.
• The differentiator: Java and Spring depth, tested harder than the algorithms. The heaviest and most distinctive bucket. Collections and HashMap internals, the Stream API and functional interfaces, exception handling, immutability, the memory model, concurrency and thread creation. A Singleton and a HashMap or stack implemented from scratch. Then Spring versus Spring Boot, REST annotations, JUnit and Mockito, and SOLID. A prepared candidate pulls ahead of an algorithm grinder here.
• The gate: Payments fluency and knowing the business. A working picture of what happens when a card is swiped or tapped, plus the payment-gateway and UPI flow. A direct behavioral bar on how Amex makes money, its closed-loop model, and which Amex products you know. A shrug of a why-Amex answer reads as a red flag. Speak to the values as themes; don't recite a list.

The plan, phase by phase

Each phase carries one focus and a checkpoint that tells you when you have earned the next. A phase is a share of your prep time, not a fixed run of weeks. The plan holds whether your runway is two months or one hard week. Watch the proportions. The online assessment leads only for campus or off-campus candidates. It is a pass-or-stop gate, so clear it cold before anything else. The medium DSA earns the technical screen. Then the plan widens sharply into Java and Spring depth. Amex spends its scrutiny there, and a strong algorithm candidate who never opened the Collections framework will stall. Practical machine coding and the payments-plus-behavioral pass close it out. Neither is filler. Don't step past a checkpoint until you can clear it without notes.

To make the proportions concrete, picture a four-week runway for an experienced hire. Skip the assessment week, since your loop opens on a live screen. Give the first days to reading the process and settling your target level against that inverted ladder. Week one goes to the medium DSA, coded in Java in a plain editor until the missing IDE stops mattering. Most of weeks two and three belong to Java internals and Spring. Cover HashMap and the memory model, a live Stream API problem, a Singleton and a hand-rolled HashMap, and a solid Spring, testing, and SQL conversation. Thread in a machine-coding drill where you stand up a running Spring Boot endpoint or a React carousel. Save the last stretch for payments design: the card-swipe walk-through and a payment-gateway sketch. Pair it with the know-Amex behavioral and a why-Amex you mean. A campus candidate shifts everything right by one block and leads with the online assessment. On a shorter runway, compress the same order without reordering it. The medium coding earns the screen, and the Java depth wins the offer.

Why the Java depth decides it

The weighting tracks where this loop departs from a generic one. At a pure algorithm shop the coding rounds are the whole filter, so you grind the hardest problems you can find. Amex works differently. Its coding bar is genuine but sits around medium and often runs in a plain editor. That rewards clean, fluent Java more than a rare trick. The scrutiny lands on the ecosystem instead. HashMap internals and the memory model, the Stream API as live code, and concurrency. A Singleton and a from-scratch HashMap, Spring and its testing tools, and the SQL underneath. Hard graph problems never build that knowledge, and an algorithm-only candidate skips it. On top sits a domain and business layer that generic FAANG prep doesn't touch. That layer covers the card-swipe flow, the payment gateway, the closed-loop model and how Amex earns. So the plan pours its widest block into Java and Spring. A fourth set of hard puzzles gets nothing, and the payments and business work counts as a gate.

The American Express angle

The trap is preparing for the wrong Amex. Spend a month on the nastiest dynamic-programming problems and you aim at a bar that tops out lower than you fear. Meanwhile the Java internals, the Spring build-and-run, and the payments story go unrehearsed, and those separate candidates. The business talk carries real weight too. Asked outright how Amex makes money, or which of its products you know, a vague answer reads as someone who didn't bother to look. Prepare for this loop specifically. You want medium DSA coded cleanly in a bare editor, and Java and Spring depth you can speak and write from scratch. You want a small Spring Boot or React build you can ship under time, plus a card-swipe and payment-gateway story you can walk. Ground your why-Amex in the closed-loop model. Spend your marginal hour on the HashMap internals you're shaky on or the payments flow you can't yet narrate. A fifth hard puzzle won't come up.

In the room

• Find your entry point before you drill. Campus and off-campus candidates clear a timed online assessment first, roughly 90 minutes and three coding problems. Some versions run 45 multiple-choice questions plus two coding tasks, or a fix-at-most-two-lines debug question. Experienced hires usually skip it and open on a live technical screen. The assessment platform is unconfirmed, so prepare for a generic timed window.
• Front-load LeetCode-medium DSA, coded in a plain editor. Arrays and strings, single-pass partitioning, matrix rotation, backtracking, and clean stack or queue implementation, written in Java without a full IDE. The difficulty is medium, so fluency and readable code matter as much as the algorithm.
• Spend your widest block on Java and Spring depth. HashMap internals, the Collections framework, the Stream API as live code, functional interfaces, concurrency, immutability, and the memory model. A Singleton and a hand-rolled HashMap from scratch. Then Spring versus Spring Boot, REST annotations, JUnit and Mockito, SOLID, and the SQL layer. An algorithm-only prep misses all of it.
• Rehearse practical machine coding you build and run. Stand up a working Spring Boot REST endpoint from an empty project. On a full-stack track, build a React carousel that fetches data and supports pause and play. Amex checks whether you can turn a short spec into running software under time.
• Bring payments fluency and a real why-Amex. Walk the card-swipe transaction end to end, and sketch a payment gateway and the UPI flow. Answer directly how Amex makes money through its closed-loop model and which Amex products you know. Pair that with a project deep dive you owned and a specific reason you want to build here.