Amazon [India] SDE II | Interview Experience
Summary
I successfully interviewed for and accepted an SDE II offer at Amazon India after a rigorous process involving online assessments, coding rounds, low-level design, high-level design, and behavioral interviews, overcoming market challenges and extensive preparation.
Full Experience
OA:
It consisted of:
- 2 Coding Questions
- System Design Questions
- Leadership Principle Questions
- Psychometric-style Assessments
Coding Questions:
- https://leetcode.com/discuss/post/6439385/amazon-sde2-oa-by-user2979x-mn77/
- https://leetcode.com/discuss/post/6445861/amazon-oa-feb-2025-by-anonymous_user-h6rb/
Solved one fully and one partially.
Followed up with the recruiter and got to know I passed. The next rounds were scheduled a few days later.
Round 1: DSA (~50 min)
Leadership Principle (LP) Questions: Answered both.
Coding Questions:
Solved both and provided optimized solutions with better space complexity.
Personal Rating: Strong Hire
Discussed my approach before diving into code, covered edge cases, and explained optimizations. Struggled a bit with space optimization in the second question but managed to solve it with just 50 seconds remaining.
Mock interviews with a friend helped a lot. It’s more about how you discuss the problem and collaboratively arrive at a solution than just solving it outright.
Round 2: LLD (~50 min)
LP Questions: Answered one and skipped one (seemed like a trap).
LLD Question:
Design a File Adapter Framework that converts one file type to another. Some adapters are already available, e.g., JSON → CSV, CSV → PDF. The goal was to create a generic framework that allows easy extension and file conversion.
Personal Rating: Neutral
Initially misunderstood the requirements and started building an input-output converter instead of leveraging the existing adapters. Realized the mistake mid-way and corrected the direction. Presented a class diagram, discussed edge cases, and suggested appropriate design patterns.
Misunderstanding the initial requirement could’ve been fatal, but scaling back quickly helped. Mock interviews definitely helped build composure under pressure. Always double-confirm requirements in LLD rounds.
Round 3 (HM): HLD (~1 hr 30 min)
LP Questions: Answered both.
HLD Question:
Design an Analytical Engine that processes IoT data from sensors and produces data streams and analytics. Later, real-time processing requirements were added.
Personal Rating: Strong Hire
This was the best round. Although I was grilled on every question, we had great discussions and even lost track of time while brainstorming system improvements. Initially unsure of the outcome, but after talking to my mentor, I realized it covered SDE3-level concepts and I did quite well.
Covered all functional requirements, and later the focus shifted to non-functional aspects like scaling for millions of sensors, identifying bottlenecks, partitioning by time/location, database choices, ETL jobs, etc.
LP questions were tough and I was also grilled on implementation-level details.
Important tip: Never fake LP answers. If you haven’t experienced something, say so. Experienced interviewers can easily detect inconsistencies and may reject you.
Round 4 (Bar Raiser):
LP Questions: Answered both.
Coding Question:
You're given a list of radiation events. Each event has a start time, end time, and a radiation value. Radiation is additive for overlapping intervals.
You’re also given a threshold. Identify all time intervals where the total radiation exceeds the threshold.
Follow-up: What if your system has only 4GB RAM and the dataset is 24GB? How would you optimize?
Personal Rating: Strong Hire
Proposed a line-sweep algorithm, covered edge cases, and implemented the solution.
For the follow-up, suggested chunking, paging, and using index structures like B+ Trees and AVL trees for efficient memory and search performance.
Leadership questions were okay in this round.
Note: Bar Raiser rounds usually happen only if there’s potential for an SDE1 or SDE2 offer.
Final Verdict:
Selected for SDE II
Accepted the offer, as I didn’t have competing offers and market conditions are rough.
For LP prep, I prepared answers for the top 30–40 questions and maintained a list of 20 key contributions from my career. Used AI to generate and fine-tune answers, and practiced speaking them out loud. Don’t over-prepare either — it can backfire. Balance is key.
I know Amazon’s culture has its flaws, but I believe with the right use of AI, work can be efficient. PIPs exist across all big tech companies now, so it’s not avoidable. My goal is to grow my expertise to Principal Engineer level while working at high intensity for a few years.
I believe I can game the stack ranking and competitive environment. Ultimately, what matters most is your manager and team. If it doesn’t work out, I can return to my previous org or join a startup.
I'm already used to high-pressure environments, late nights, and on-calls — so I'm prepared. AI made my life much easier.
Journey:
Currently working at a mid-level startup. Landed this offer after ~3 months of prep.
Key takeaway: Don’t give up. Keep grinding until you break through.
The hardest part is getting the interviews. I sent over 600+ applications, cold emails, and referrals. For some companies, direct applications worked. Others contacted me via LinkedIn.
Referrals didn’t work for me. Even people I know personally were unwilling to help.
Be self-reliant. Find your inner strength.
Had also reached the final round at Qualcomm for a Data Engineering role with strong positive feedback — but the recruiter ghosted me.
That hit hard.
My Amazon Bar Raiser was scheduled around the same time and I felt completely drained. Stopped practicing.
But then — fate intervened. My headphones stopped working, the round got delayed by a week, and I used that time to regroup and prepare.
If it’s meant to be, it will happen. Believe in destiny.
Also had interviews with Slice, Meesho, DigitalOcean, and others. Got rejected or ghosted due to minor prep gaps or just stronger candidates.
Atlassian and DE Shaw rejected me outright for not having Java experience despite showing interest. Shows how saturated the market is. COVID-era hires,Tech Influencers Glorifying SE roles, AI-based layoffs, and an oversupply of CS grads have made it worse.I can only imagine how hard it is for the freshers right now, I can feel their pain. 2026 might be even tougher.
Some Advice:
- Avoid LinkedIn/Youtube influencers and Topmate "gurus" selling overpriced courses or vague advice.
Most are just profiting off desperate job seekers. - Yes, my Tier-1 college and some strong projects helped, but you can offset that with standout personal projects, especially in trending areas like Agentic AI.
- Years of experience and strong fundamentals can compensate for pedigree.
My Strategy?
Grow up.
Why would anyone publicly share a strategy that works? And even if it did, it wouldn’t be effective once everyone starts copying it. This is the same thing i want you to avoid.
Instead, study LeetCode/Reddit threads, use AI tools smartly, and make personalized prep plans. Forge your own path.
Don’t cheat. No shortcuts. Karma catches up sooner or later.
Closing Thoughts:
Interviews aren't just about evaluation — they’re a journey of self-discovery.
Work on soft skills. Software engineering is a team sport, not a solo mission. Make sure you are able to drive the narrative to your advantage, collaborate cross functionally. Build trust, make connections, and figure out what drives you — Money, Power, Promotions, or Impact?
I’m personally driven by Impact and Power.
Think long-term. Specialize in a Niche. Build depth and breadth. Most prep resources are already out there — pick the one that clicks for you.
Practice > Watching/Reading
I am incredibly thankful to the Leetcode and Reddit community, and I will try to give back to the community somehow.
Pro Tips:
- Emergency Funds: Save 6 months’ worth of expenses and also think long term.(stocks, MFs, gold, FDs, etc.)
- Avoid EMIs and loans before getting health and term insurance.
- Invest in learning (new domains and skillsets) and health (e.g., sports, gym — I swim).
- Leverage trends (e.g., GenAI, Agentic) to pivot quickly if needed to a different role.
- Never leave a job without an offer in hand.
It’s a dog-eat-dog world — but always stay kind.
A little empathy goes a long way.
Be grateful. Help others. Thank you for reading this far.
Taking a short break before returning to the All grind LeetCode Hotel.
Interview Questions (5)
Merge K Sorted Lists
Product of Array Except Self
Given an integer array nums, return an array answer such that answer[i] is equal to the product of all the elements of nums except nums[i].
Design a File Adapter Framework
Design a File Adapter Framework that converts one file type to another. Some adapters are already available, e.g., JSON → CSV, CSV → PDF. The goal was to create a generic framework that allows easy extension and file conversion.
Design an IoT Analytical Engine
Design an Analytical Engine that processes IoT data from sensors and produces data streams and analytics. Later, real-time processing requirements were added. The discussion covered all functional requirements, and later shifted to non-functional aspects like scaling for millions of sensors, identifying bottlenecks, partitioning by time/location, database choices, ETL jobs, etc.
Radiation Events with Threshold
You're given a list of radiation events. Each event has a start time, end time, and a radiation value. Radiation is additive for overlapping intervals. You’re also given a threshold. Identify all time intervals where the total radiation exceeds the threshold. Follow-up: What if your system has only 4GB RAM and the dataset is 24GB? How would you optimize?
Preparation Tips
I prepared for approximately 3 months. My preparation included conducting mock interviews with a friend to build composure and improve collaborative problem-solving. For Leadership Principles, I prepared answers for 30–40 common questions, maintained a list of 20 key career contributions, used AI to generate and fine-tune my responses, and practiced speaking them aloud. I studied LeetCode and Reddit threads, utilized AI tools smartly, and created personalized prep plans. I also focused on developing soft skills and emphasized active practice over passive learning methods like watching or reading.