High-Score (Bugfree) Amazon SDE-2 Interview Experience

A concise write-up of my Amazon SDE-2 interview loop — applied via referral with ~2 years of fintech full-stack experience. The process included an online assessment, a pause by the recruiter due to location, reconnection for Hyderabad onsite scheduling, a 4-round in-person loop, and a virtual bar-raiser. Final status: selected.

Quick timeline

• Applied via referral (2 years fintech, full-stack)
• Online assessment (OA): straightforward
• Recruiter paused because of location, later reconnected for Hyderabad
• In-person loop (3 rounds) + virtual bar-raiser (4th round)
• Final: selected

---

Interview breakdown (round-by-round)

Round 1 — LP + Coding

• Leadership Principles (LP) behavioral questions
• Coding problems: "Max Points from Cards" and "Asteroid Collision"
• Notes: Did a dry run on Asteroid Collision, caught a bug, fixed it during the interview — good debugging demonstration

Round 2 — LP + Low-Level Design (LLD)

• LP discussion
• LLD: Tic-Tac-Toe
  • Covered required entities, class design, and class diagram
  • Implemented the game-over check method in C++

Round 3 — LP + High-Level Design (HLD)

• LP behavioral questions
• HLD: Event Booking System (architecture, components, scalability considerations)

Round 4 (Bar Raiser — virtual) — LP + Coding

• LP-focused behavioral assessment by the bar raiser
• Coding problems: "House Robber II" and "Boats to Save People"
  • House Robber II: circular DP variant — deal with circular constraint by splitting into two linear runs
  • Boats to Save People: two-pointer greedy approach to pair lightest and heaviest

---

What worked well

• Clear LP stories: used STAR format (Situation, Task, Action, Result) and tied examples to Amazon's principles
• Dry runs: doing a hand-simulation helped catch a bug live (shows attention to detail)
• LLD/ HLD clarity: focusing on entities, APIs, data flow, fault tolerance, and scaling trade-offs
• Language comfort: implemented C++ method when asked — be comfortable in at least one language

Quick tips if you're preparing for a similar loop

• Prepare 6–8 strong LP stories mapped to Amazon's Leadership Principles; rehearse concise STAR answers
• Practice dry runs for coding problems — simulate inputs/outputs and edge cases on paper
• For LLD: sketch entities, relationships, responsibilities, and one or two key methods
• For HLD: discuss capacity, data partitioning, caching, consistency, and failure scenarios
• Know common patterns: two-pointers, stacks, DP (including circular DP), and greedy strategies

---

Final notes

Recruiter logistics questions (compensation, notice period, location) were asked toward the end and I accepted the offer. The combination of strong LP storytelling, careful dry runs during coding, and clear design explanations helped secure the role.

Good luck — and debug your dry run!

#AmazonInterview #SystemDesign #SoftwareEngineering