Placement Prep · Product companies
Amazon SDE intern & new-grad prep: patterns, not problem counts
Product-company interviews, Amazon’s included, test whether you can recognise a pattern, explain a solution and code it cleanly while someone watches. Solving 800 problems without the patterns rarely gets you there. This guide maps the patterns that typically appear in SDE intern and new-grad loops to free labs, gives a 6-week plan, and ends with a mock-interview checklist.
Free interactive DSA labs · see each pattern run before you code it
How the process typically runs
Details vary by role, location and year, so rely on what your recruiter tells you. Commonly reported stages for SDE intern and new-grad roles are:
- Online assessment: usually a couple of coding problems in a time limit, often with a work-style or behavioural questionnaire and sometimes a short reasoning section.
- Interviews: one or more rounds, each typically mixing a data-structures and algorithms problem with behavioural questions.
- Behavioural questions are commonly framed around Amazon’s publicly published Leadership Principles, so prepare real stories from projects, internships and college work.
The pattern map: from problem shape to lab
When you read a problem, ask “what shape is this?” before “how do I code it?”. These patterns cover a large share of what new-grad interviews ask.
| Pattern | Signal in the problem | Practise it |
|---|---|---|
| Hash map | “Have I seen this before?”, counts, pairs | Hashing |
| Two pointers | Sorted array, pairs, in-place rearranging | Two pointers |
| Sliding window | Longest/shortest contiguous sub-array or substring | Sliding window |
| Binary search | Sorted data, or “minimum X such that” | Binary search |
| Monotonic stack | Next greater/smaller element, spans, histograms | Monotonic stack |
| Linked lists | Reverse, merge, detect a cycle, find the middle | Linked list |
| Trees | Traversals, depth, lowest common ancestor, BST checks | Binary tree, Recursion |
| Heap / top-k | K largest, K closest, merge K sorted, scheduling | Heap, Quickselect |
| Intervals | Overlapping meetings, merging ranges | Merge intervals |
| Graph BFS/DFS | Islands, shortest steps, connected parts of a grid | BFS/DFS, Grid BFS |
| Dependencies | Prerequisites, build order, detect impossible orders | Topo sort |
| Connectivity | Merge accounts, redundant connection, groups | Union-Find |
| Backtracking | Generate all subsets, permutations, combinations | Backtracking, N-Queens |
| Dynamic programming | Count ways, min cost, best subsequence | Coin change, DP table |
| Design + DSA | Build a cache or a small data structure with O(1) operations | LRU cache, Trie |
The 6-week plan
About 3 hours a day. The rule for every week: learn the pattern in the lab first, then solve 3 to 5 problems that use it, re-solving one from memory a day later.
| Week | Patterns | Also do | Labs |
|---|---|---|---|
| Week 1 | Complexity, hash maps, two pointers, sliding window | Pick your interview language; learn its standard library cold | Big-O, Hashing, Sliding window |
| Week 2 | Binary search, sorting, monotonic stack, intervals | Write 3 behavioural stories | Binary search, Monotonic stack, Intervals |
| Week 3 | Linked lists, recursion, trees | First mock interview with a friend | Linked list, Binary tree |
| Week 4 | Heaps, graphs (BFS, DFS, topological sort, Union-Find) | Timed online-assessment practice: 2 problems, fixed time | Heap, BFS/DFS, Topo sort |
| Week 5 | Backtracking and dynamic programming | Two more behavioural stories; second mock | Backtracking, Coin change, DP table |
| Week 6 | Mixed review; design-flavoured problems like an LRU cache | Two full mocks; revise your mistake log | LRU cache, Readiness check |
Practise this now
Sliding Window
Best sub-array of size k in one pass: add what enters, subtract what leaves. O(n·k) becomes O(n).
Binary Search
Halve a sorted array every step, plus the two off-by-one bugs that cost marks under time pressure.
Monotonic Stack
Next greater element and stock-span in O(n): keep a stack that only ever grows in one direction.
Binary Trees
Inorder, preorder, postorder and level order on one tree, plus height and BST checks.
Heap & Priority Queue
K largest, merge K sorted lists, scheduling. Watch push and pop travel one O(log n) path.
Graph Search: BFS vs DFS
BFS floods outward in rings and finds the shortest path; DFS dives deep. The queue-vs-stack difference.
DP Table Builder
LCS and edit distance: watch a 2-D table fill cell by cell from smaller sub-problems.
LRU Cache
Hash map plus doubly linked list gives O(1) get and put. A favourite design-meets-DSA question.
Mock-interview checklist
Run every practice problem through these steps, out loud. Interviewers typically evaluate how you think, not only whether the code passes.
- Clarify. Ask about input size, duplicates, negatives, empty input and the expected output format.
- Work an example by hand, including one edge case.
- State a brute force and its time and space complexity.
- Name the pattern and explain how it improves the brute force.
- Code it cleanly with clear variable names, talking as you go.
- Test it by walking your example through the code line by line.
- Finish with complexity and one possible follow-up improvement.
For behavioural rounds, prepare 6 to 8 true stories in a Situation, Task, Action, Result shape, each with a concrete result and what you learned.
Frequently asked questions
What is asked in the Amazon SDE online assessment?
It has commonly included a couple of coding problems in a time limit, along with a work-style or behavioural questionnaire. The exact format varies by role and year, so follow the instructions in your assessment invitation.
How many LeetCode problems do I need for Amazon SDE?
There is no magic number. Candidates who learn the 12 to 15 core patterns well and can explain their solutions out loud tend to do better than those who solve many problems without recognising the underlying pattern.
Which DSA topics are most important for SDE intern and new-grad interviews?
Arrays and hash maps, two pointers, sliding window, binary search, linked lists, trees, heaps, graph traversal, and some dynamic programming and backtracking. Design-flavoured problems such as an LRU cache also appear.
How should I prepare for behavioural questions?
Write 6 to 8 true stories from your projects, internships and college work in a Situation, Task, Action, Result shape. Amazon publishes its Leadership Principles, so map each story to the principles it demonstrates.
How long does it take to prepare for a product-company SDE interview?
With one programming language already comfortable, around six weeks of daily pattern practice plus regular mock interviews is a realistic plan. Starting from scratch, add a few weeks of fundamentals first.
Is BitBootcamp affiliated with Amazon?
No. BitBootcamp is an independent learning platform. The company name is used only to describe the interviews this guide helps you prepare for.
See how interview-ready you are today.
Take the 2-minute readiness check, then follow the placement path from DSA core to system design.
All prep guides TCS NQT Infosys Wipro · Accenture · Cognizant DSA in Hindi
BitBootcamp is an independent learning platform and is not affiliated with, endorsed by, or partnered with any company named on this page. Company names are used only to describe the recruitment drives students prepare for. Exam patterns change between seasons, so always confirm the current format in the official notice for your drive.