American Express is the credit card and travel services giant — large engineering organization with significant tech investment. The interview is rigorous for a financial-services company, with emphasis on payments, fraud detection, and the regulatory complexity of card issuing.
Process
Recruiter screen → 60-minute coding phone (DSA medium) → onsite virtual: 2–3 coding, 1 system design, 1 behavioral. Cycle: 4–6 weeks.
What they actually ask
- Design a real-time fraud detection system at sub-100ms latency. Interviewers probe how you score a transaction in-line without blowing the latency budget — expect questions on fast feature lookups from a low-latency store, model serving, and what you do when the model call times out (fail open and approve, or fail closed and decline). Be ready to discuss stream processing, rolling-window aggregates over a card’s recent activity, and how you trade off false positives against blocking legitimate spend.
- Design a points/rewards calculation engine. Focus on correctness and idempotency — points must be awarded exactly once even if the transaction event is delivered twice, so talk through a deduplication key and how you reconcile against the ledger of record. Interviewers like to see you handle category multipliers, limited-time promotions, and back-dated adjustments without double-counting.
- Design payment processing with regulatory compliance. Walk through the authorization → clearing → settlement flow and where idempotency and audit logging live, since every state change on money has to be reconstructable for auditors. Expect follow-ups on PCI-DSS scope (never storing raw card numbers), retries against a flaky downstream network, and exactly-once guarantees on money movement.
- Coding: medium DSA, often with financial framing. Problems are standard medium difficulty but wrapped in payments language — a “merchant transaction stream” is still a hash-map or heap problem underneath. Practice the 15 core patterns across arrays, strings, hash maps, and graphs, and narrate your Big-O out loud so the interviewer sees you weighing time against space.
- Behavioral: customer focus, ownership, navigating regulated environment. Prepare STAR stories that show you put the cardmember first and took end-to-end ownership of a problem, plus at least one about working within compliance constraints rather than around them. A strong example describes a time you slowed a launch to get a control right and still shipped on a reasonable timeline.
Levels and comp (2026)
- SE: $135K–$170K total
- Senior SE: $200K–$260K
- Staff: $290K–$380K
- Principal: $400K–$520K
Prep priorities
- Be fluent in Java (the bulk of the codebase) or Python (newer services)
- Understand payments protocols and card networks. Know the roles in a card transaction — cardholder, merchant, acquirer, network, and issuer — and the auth-to-settlement lifecycle, since AmEx is unusual in acting as both the network and the issuer. You don’t need ISO 8583 byte layouts, but you should be able to explain what happens between a card swipe and the money landing in the merchant’s account.
- Brush up on regulatory framework: PCI-DSS, OCC, BSA/AML. You won’t be quizzed like a compliance officer, but you should know what each one governs — PCI-DSS for cardholder data security, BSA/AML for detecting money laundering — and be able to name one way it shapes system design, such as encryption-at-rest or transaction monitoring. Treating regulation as a design input rather than a nuisance goes a long way in these rooms.
Frequently Asked Questions
Is AmEx remote-friendly?
Hybrid in NYC (HQ), Phoenix, others. 3 days in office for most roles.
How does AmEx compare to Capital One or Goldman Sachs?
Capital One is more tech-forward. Goldman pays more for top engineers. AmEx has strong premium-card brand and stable benefits.
What is the engineering culture?
Mature, methodical, regulation-aware. Less startup-fast than tech-native companies. Stable employer with strong benefits.
Similar company guides
Prepping for American Express? Put it to work:
