Amazon New Grad Interview Experiences (2026)

🌱 18 recent Amazon New Grad interview experiences are collected below, ordered from newest (Sep 4, 2026) to oldest (Apr 25, 2026). Matching interview questions link directly to Amazon practice on FastPrep (p.s. sadge just found question links are not working for now, i am fixing them.

Behavioral, Leadership Principles, and Resume Questions from Reported Interviews

Ownership and initiative

1. Work outside normal responsibilities.
2. A team heading in the wrong direction and the action taken.
3. An improvement driven without being asked.
4. Stepping outside a comfort zone.
5. Learning a technology and applying it to a real project.

Judgment, ambiguity, and tradeoffs

1. A difficult decision or important tradeoff.
2. Sacrificing a short-term benefit for long-term success.
3. Moving forward with ambiguous requirements or missing data.
4. An earlier solution that would be reconsidered today.

Delivery, obstacles, and failure

1. A tight deadline, major obstacle, or project blocker.
2. A mistake with real consequences and how it was handled.
3. Investigating a difficult problem and finding its root cause.
4. What would be done differently in hindsight.

Feedback, trust, and conflict

1. Negative feedback or criticism and the response.
2. Conflict with a teammate or a difficult teammate.
3. Disagreeing with the team's direction.
4. Collaborating outside the immediate area.

Motivation and reflection

1. Why Amazon?
2. Which project best represents the way the candidate works?
3. What was hardest about the most recent project?
4. What important achievement has not yet come up?

GenAI

1. Using GenAI in a project or development workflow.
2. When GenAI is and is not a good fit.
3. Unexpected AI output or a real hallucination and the response.
4. Validating AI-generated output.
5. Improving a GenAI feature's reliability and accuracy.
6. Useful or interesting GenAI applications.

Common follow-ups: What did you personally own? Why that approach? What alternatives, data, and tradeoffs did you consider? How did others respond? What measurable result followed? How did you verify the root cause? What would you do differently now? Which Leadership Principle does the story demonstrate?

Preparation Formula

DSA + Leadership Principles + LLD/OOD + Resume Depth + Testing and Follow-Ups

For 2026 loops, add GenAI Fundamentals + Practical AI Use + Responsible Validation of AI Output.

18 Recent Amazon New Grad Interview Experiences

01. United States: LRU Cache and Stock Trading DP

Post date: September 4, 2026

Four rounds were scheduled, but one was canceled, leaving three interviews for an SDE I New Grad role.

1. Leadership Principles, then LRU Cache with extra inputs and extensions.
2. Leadership Principles, then a data structure design task for determining the relationship between two people. The relationship model was not specified.
3. GenAI, then the dynamic programming problem Best Time to Buy and Sell Stock with Cooldown. Coding was included even though a behavioral-only round had been expected.

02. Train Fares, Strings, and a Sliding Window

Post date: September 3, 2026
Outcome: Rejected

1. Train Route Fare Calculator: for G → U → H → K → I, each segment cost 1; API: calculateFare(start, stop). Follow-ups modeled stations/routes as objects and capped weekend fares. Recognizing the OOD requirements early was the challenge.
2. An unspecified string/substring problem with tests, including an empty string, plus GenAI.
3. Behavioral questions about difficult teammates and decisions, with detailed follow-ups.
4. Implement SlidingWindow(k), put(number), and getMax() for the latest k values. The solution used O(k) space and O(k) getMax(); follow-ups compared arrays, hash maps, linked lists, heaps, and a monotonic queue. Practice Sliding Window Maximum and the stateful Sliding Window Maximum API.

03. Top K, Shopping Cart Design, and an Iterator

Post date: August 31, 2026

Three video interviews lasted about one hour each.

1. A Top K frequency problem.
2. An object-oriented design question about a shopping cart.
3. Implementing an iterator.

The shopping cart requirements and iterator type were not specified, so the linked shopping cart exercise is a reference practice version.

04. Vancouver NG26: In-Person Interview Loop

Post date: August 31, 2026

The four rounds were shared out of order. Amazon covered flights and a hotel for two days, plus $70 per day for meals and $100 for transportation; the currency and allowance period were unspecified.

1. LLD and GenAI about past projects, suitable use, interesting applications, and reliability. Build a linked-list queue with O(1) enqueue/dequeue without native data structures; add delete(x) and removeAllDuplicates().
2. Two LP questions, then merge two dictionaries with loops and extend to k dictionaries. Merge rules were unspecified.
3. Behavioral-only Bar Raiser using internship stories.
4. One BQ, then Median of a Data Stream, framed as an API. Also practice the LLD version.

05. Trees, a Monotonic Stack, Pizza Design, and Meeting Rooms

Post date: August 28, 2026

Technical topics: a tree problem using height, DFS, and LCA; monotonic stack; Pizza OOD; and meeting rooms with a priority queue. Related pattern practice: Lowest Common Ancestor Implemented with a Stack and Meeting Rooms II. Exact prompts were unavailable.

Behavioral questions covered common LP topics and AI output; follow-ups tested whether each story demonstrated the relevant principle.

06. Rate Limiter, Top K, and Spreadsheet Design

Post date: August 27, 2026
Outcome: Rejected

1. Rate limiter. Similar practice: Logger Rate Limiter (LC 359) and Design Hit Counter (LC 362).
2. AI usage, then catalog/Top K with memory optimization. Related practice: LC 215 and Kth Largest Element in a Stream (LC 703).
3. Behavioral-only Bar Raiser on missing data, conflict, and root cause; follow-ups examined decisions, alternatives, evidence, and results.
4. Spreadsheet cell dependencies. LC 631 and LC 3484 were similar practice, not exact copies.

07. Seattle AWS/AUTA: Four Rounds and an Offer

Interview dates: August 20 and August 24, 2026
Post date: August 29, 2026
Outcome: Offer on August 27, 2026

1. Behavioral questions about resolving a blocker and what could be done differently, then Path Sum II and an unspecified harder variation. Hints helped with the follow-up.
2. A question about work outside normal responsibilities, then an AWS Billing Service for one user based on services, usage, and rates. Discussion focused on classes, choices, tradeoffs, and changing requirements.
3. About 30 minutes of GenAI discussion, then First Non-Repeating Character and an unspecified Merge Intervals variation.
4. A behavioral-only Bar Raiser with detailed questions about decisions and rejected alternatives.

08. Vancouver: User Logs, DP, and Playlist Design

Interview date: August 24, 2026
Post date: August 29, 2026
Outcome: Rejected three days later

Three onsite rounds and one virtual round happened that day.

1. From chronological access logs, find the earliest user who appears exactly once, then discuss complexity and LPs.
2. An unspecified dynamic programming problem, followed by everyday AI use and a real AI hallucination.
3. LPs, projects, and learning, then playlist-generator LLD with many follow-ups. Playlist Source Mixer is related practice; its mixing rules were not confirmed requirements.
4. BQs about conflict, leaving a comfort zone, recent projects, and project difficulties.

09. Reward Points, String Rearrangement, and Movie Ticket Pricing

Post date: August 21, 2026
Outcome: Process marked as passed; no offer update

1. A behavioral-only Bar Raiser.
2. Pay with the fewest reward points. Each program had a name, balance, and changing cash value. Example: cover $168 using 36,000 airline points at $0.012, 12,000 store points at $0.009, and 25,000 card points at $0.0085. Respect balances; fail if payment is impossible. Overpayment and fractional points were unspecified; the linked exercise defines them.
3. Rearrange a string so adjacent characters differ. Return any valid result or failure. The linked judge uses deterministic ordering.
4. Extensible movie ticket pricing. Format sets base price; showtime and seat tier apply multipliers; extras have fixed prices. Price = (base + extras) × showtime × seat tier. Support new options without major changes.

10. Front End Engineer NG26: Phone Screen and Final Rounds

Interview dates: July 16, August 13, and August 19, 2026
Post date: August 22, 2026
Timeline: Applied June 2; scheduled July 2–8; phone screen July 16; advanced July 28; two final interviews August 13; Bar Raiser August 19. The team was AWS EC2 Pricing, with a possible later move toward full-stack work.

1. Phone screen: BQs on self-directed improvement, extra responsibility, and hindsight. Recreate a pictured UI with three numbered buttons, each opening a closable floating window with fetched data. The editor had no autocomplete or documentation.
2. Hiring manager: “Why Amazon?”, then render nested comments hierarchically from IDs, parent IDs, and text. An N-ary tree was implemented; much of the 30 minutes went to clarification. BQ: a consequential mistake and its handling.
3. Senior engineer: Project and technology choices; JavaScript let vs. var; layout centering; then an Amazon-style search bar returning the ten most relevant products. Follow-ups moved from hardcoded data to an API and debouncing. Marketplace Product Search Service covers broader service design; the interview focused on frontend work.
4. Bar Raiser: Deep investigation; correcting a team's direction; learning and applying frontend technology, including responsive devices; and anything important not previously covered.

11. Phone Interview: Behavioral, GenAI, and Coding

Post date: August 18, 2026

Behavioral questions covered a tight deadline, criticism, and ambiguous requirements. GenAI covered AI experience and chatbot architecture. Stateless Generative AI Chat Service is related practice; detailed interview requirements were unavailable.

Coding covered Valid Parentheses, Two Sum, and an unspecified Two Sum variation.

12. Offer: File Search, Caching, and Building a Tree

Interview dates: July 20 and August 12, 2026
Post date: August 17, 2026
Outcome: Offer on August 17, 2026

Timeline: VO survey July 8; three consecutive interviews July 20; final Bar Raiser August 12; offer August 17.

1. Behavioral-only round covering Dive Deep, negative feedback, and GenAI.
2. BQ on sacrificing short-term benefit for long-term success. Build a Unix file search API for filenames under a directory; add minimum size, AND filters, and symlink safety. Related LLD: File System Search with Symlink Safety.
3. BQ on an obstacle. Design an LRU query-result cache for a multi-time-zone database with traffic peaks. Follow-ups: get() complexity, concurrency failures, sequential IDs 1, 2, 3, and rising traffic.
4. BQ on leaving a comfort zone. Solve Create Binary Tree from Descriptions (LC 2196). Follow-ups: hash-map value, complexity, and out-of-order descriptions that initially form a forest.

13. Parking Lot Design, Graphs, and Binary Search

Post date: July 27, 2026
Outcome: Pending when posted

1. Leadership Principles, resume, and a project deep dive.
2. Two medium/hard coding questions on linked lists and graphs. Both were completed, with some guidance on the second; exact prompts were unavailable.
3. Parking lot LLD with unspecified requirements. Related practice: Design a Multi-Floor Parking Lot.
4. Hiring-manager and GenAI questions about AI use, then an unspecified binary-search-on-the-answer problem.

14. Amazon NG 2026: Four-Round VO

Interview date: July 23, 2026
Post date: August 14, 2026
Outcome: Rejected

Timeline: Applied in January; VO invitation July 7; four interviews July 23.

1. GenAI BQ about unexpected output, a concrete example, and deep follow-ups. Find the Top K slowest log entries with a priority queue; it was described as similar to LC 636, but the full prompt was unavailable.
2. BQ about work outside the normal responsibility area. Design a package dependency manager and return a valid installation order using topological sorting.
3. BQs about tradeoffs and a challenge. Solve Group Anagrams (LC 49) first with a hash map, then discuss a bucket-based optimization.
4. Behavioral questions about a tight deadline and teammate conflict.

15. In-Person Final: Server Assignment, Loyal Customers, and Rate Limiting

Post date: July 8, 2026
Outcome: Positive result, but no offer because of current business needs

The US SDE I New Grad interview took place in the first week of July.

1. LPs, then assign partitions to the least-loaded server. A min heap was implemented; follow-ups compared a greedy alternative and optimization.
2. LPs, then use two days of logs to find customers present both days with at least two unique products overall, using nested dictionaries and sets.
3. A behavioral-only Bar Raiser.
4. Design an extensible log rate limiter with separate per-message and global policies. Each policy initially mutated state even when another rejected the event; the design was revised so state matched the final decision.

16. Cash Change, Linked List Cycle, and Caesar Cipher

Interview date: June 18, 2026

Four back-to-back interviews covered:

1. BQs on extra responsibility and a commitment at risk. Return minimum cash pieces using $20, $10, $5, $1, 25¢, 10¢, 5¢, and 1¢; 6.35 can be $5 + $1 + 25¢ + 10¢. Follow-ups added early exit and limited inventory with exact-change failure. Practice: Minimum Cash Pieces with Limited Inventory.
2. BQs on deep analysis, influencing disagreement, improving after feedback, and later applying that learning.
3. Detect and break a linked-list cycle by setting the offending next to NULL. Follow-ups traced Floyd's pointers, explained correctness, found the entry, and compared a hash table. GenAI covered an efficiency use case and validation; resume depth covered memory-ordering for IPC.
4. BQs on the hardest work, personal contribution, and debugging. Implement a Caesar Cipher with a Precomputed Alphabet: preserve case, wrap letters, and keep nonletters unchanged. Normalize the shift once and precompute 26 mappings.

17. Permutations, Tree Relationships, and Pub/Sub

Interview date: May 13, 2026
Post date: May 14, 2026
Outcome: Not specified

1. Behavioral questions and follow-ups only.
2. After BQs, print every ordering of N students, then arrange them in a circle: Circular Student Arrangements. Next, classify two tree nodes as siblings, cousins, or neither.
3. Resume, fundamentals, and an unspecified Longest Consecutive Sequence variation in which adjacent numbers could differ by less than k.
4. Behavioral questions, then object-oriented design for an in-memory publish/subscribe system.

18. BFS, Meeting Rooms, and a Deck of Cards

Post date: April 25, 2026
Outcome: Rejected

1. BQ about GenAI experience, then an unspecified BFS problem with multiple starting points.
2. BQs about obstacles, several tasks under a tight deadline, and revisiting a solution and its tradeoffs. Coding covered Meeting Rooms and Meeting Rooms II.
3. BQ about proactively working outside normal responsibilities, then shuffle and sort a deck of cards in place. Suit order was Clubs, Diamonds, Hearts, Spades; ranks ran from Ace through King.

These experiences were gathered from LeetCode Discuss, Reddit, Glassdoor, and just about anywhere else useful interview information could be found. Come hang out on FastPrep Experience and share whatever’s on your mind. And a special thank you to a special friend for sharing their experience in this roundup! 🧡