Amazon — Interview Playbook (SDE / Senior orientation)

Amazon’s loop is famous for Leadership Principles (LPs) and behavioral depth alongside solid technical rounds. Your recruiter lists the exact schedule; use this as study framing only.

Leadership Principles — how to use them

• Official list: rely on Amazon’s current published LPs; names and emphasis occasionally evolve—verify on amazon.jobs or materials your recruiter sends.
• Behavioral interviews are often LP-scoped: prompts map to principles such as Customer Obsession, Ownership, Dive Deep, Deliver Results, Have Backbone; Disagree and Commit, Insist on the Highest Standards, Earn Trust, etc. (confirm current set).

Mapping stories without inventing detail

• Build a matrix: rows = your real situations (projects, incidents, conflicts); columns = LPs each story can illustrate.
• One story can support multiple LPs if you emphasize different facets honestly (e.g., ownership for scope, dive deep for debugging—only if true).
• Do not fabricate metrics, titles, or incidents. If an interviewer asks for numbers, say what you can share or describe impact qualitatively.
• Prefer STAR (see Behavioral STAR guide) with clear “I” actions versus vague “we.”

Technical rounds — common expectations

Coding

• Medium difficulty is typical for many SWE loops; senior loops may add harder follow-ups or design-flavored problems.
• Strong signals: working solution, clean complexity discussion, edge cases, and structured debugging if something is wrong.
• Amazon-specific flavor often includes operational thinking: how would this run in production (timeouts, bad input, throttling) when the problem invites it.

System design (level-dependent)

• End-to-end thinking: requirements, API, data stores, scaling paths, failure handling.
• Cost and operability are valued; connect choices to customer impact and operational burden when relevant.

Bar raiser / LP-aware technical

• Some interviewers probe how you made decisions (tradeoffs, escalation, customer focus) inside technical discussions—keep answers factual.

Preparation checklist

• LP stories: at least two strong stories per common LP cluster you expect to face; each story should survive five follow-up questions (timeline, data, tradeoffs, conflict).
• Coding: consistent practice on arrays/trees/graphs/intervals and reading constraints carefully (Amazon problems often reward precision).
• Design: practice one retail-adjacent or high-scale consumer scenario only if it matches your target team; otherwise stick to universal primitives from this repo’s design track.

Pitfalls to avoid

• Keyword stuffing LPs without substance—interviewers cross-check with follow-ups.
• Blaming former teams or managers.
• Vague ownership: clarify your role without overstating it.

Day of

• Bring silence-tolerant structure: “Let me repeat the problem, list edge cases, then code.”
• For behavioral: answer the LP they asked for; if a story fits multiple, name which angle you are using first.

Bar Raiser concept (public framing)

Amazon publicly describes a Bar Raiser role focused on long-term hiring standards. You may never know which interviewer holds that hat. Practical takeaway: treat every round as calibration-quality—complete answers, explicit tradeoffs, and honest scope of your contribution.

LP drill without sounding scripted

• For each prepared story, write three follow-up questions an interviewer might ask and answer in two or three sentences each.
• Include at least one prompt about failure, one about disagreement, and one about data or customer signal—these map cleanly to several LPs when true for your experience.

Technical depth by round type

Pure coding

• Precision on empty, single element, and overflow edge cases.
• Complexity stated after a correct solution; refine if the interviewer asks for better.

Coding + operational extension

• When the prompt maps to a service (rate limit, debounce, cache), outline API, failure, and monitoring briefly—only as much as time allows after a correct core.

System design

• Customer-facing SLA thinking: what must never break vs what can degrade gracefully.
• Operational cost: human on-call load, runbooks, noisy alerts—tie to ownership and customer trust when genuine for your background.

Design structure helper: System design interview framework (RESHADED)

STAR reminders for Amazon behavioral

• Situation minimal; Action heavy with your verbs (defined, drove, measured, escalated).
• Result includes business or customer angle when possible; learning outcome counts when launches did not land.

Full framework: Behavioral STAR guide

Logistics and tone

• Clarify levels and scope of the role with your recruiter; loop difficulty tracks target level.
• Customer Obsession does not require retail examples—internal customers and developer productivity count when authentic.

Anti-patterns in LP answers

• Hero narrative with no team context—interviewers ask for collaborators and constraints.
• Perfection claims—credible stories include mistakes and adjustment.
• Jargon from blog posts that never appeared in your actual decision record.

Study pairing with this repo

• Use 12-week study roadmap for topic sequencing; add one LP journaling session weekly (short bullets: situation → LP → outcome) using only real episodes from your career.