Pass the FAANG interview.
Without grinding LeetCode.
You learn how to solve Two Sum. You never learn how to recognize that a problem wants a hashmap. So you grind 300 problems — and still freeze on novel ones. FaangDeck teaches the recognition first.
Why 217, not 600? Every problem in Blind 75 + NeetCode 150 + Grind 75 — taught, not listed. 217 is the set that actually teaches the 19 patterns. The other 400 are reps.
300 problems and still stuck on a new one. Sound familiar?
Memorizing solutions doesn't transfer. Learning to recognize the pattern does.
How most people prep
- Solve LC-1, LC-2, LC-3 in order
- Grind 300 problems hoping patterns emerge
- Freeze on LC-301 — can't tell which technique to reach for
- Memorize specific solutions, not transferable signals
- 3–6 month timeline. Most quit.
How FaangDeck teaches
- Recognition first. Three signal slides per lesson — the verbal cues in the problem statement that point to one technique.
- Templates second. The canonical Python skeleton you adapt for every variant of that pattern.
- Traps third. The near-correct bug everyone hits, surfaced deliberately rather than hidden.
- Active recall. 1,503 quiz questions and 1,052 flashcards, baked in — not bolted on.
- Interview communication. Every deck closes with two slides on what to say in the room.
Other things you'd expect from a real product:
Open the app. Pick a lesson. Done.
No signup, no setup, no tutorial. Here's the actual flow.
Open the app
One tap on the link. Lands on Browse → 6 phases → 19 topics → 217 lessons. No account, no signup, no email gate.
Read the lesson
~10 minutes per problem deck. Hook → recognition signals → template → solution → trap → complexity → interview playbook. Same structure every time.
Quiz yourself
Multiple-choice questions test pattern recognition + template recall. Tap ❓ at the end of any deck. Wrong-answer explanations on every question.
Flashcard the gaps
1,052 active-recall cards across the curriculum. Tap 🎴 to drill. Spaced review surfaces the patterns you're shakiest on.
Every pattern an interviewer can throw at you.
19 topics in 6 phases, in learning order. Each starts with a master overview, then problem deep dives, then a review.
Every problem in Blind 75, NeetCode 150, and Grind 75 — plus prerequisite scaffolding before them and advanced extensions after.
Covers Blind 75, NeetCode 150, and Grind 75 — every problem, taught in the same pattern-first arc.
Blind 75
The original 75-problem list curated by an ex-Facebook engineer. The minimum surface area every FAANG candidate is expected to recognise. All 75 are in FaangDeck, distributed across the 6 phases — visitors hit them naturally as they progress.
NeetCode 150
The 150-problem expansion organised by pattern. The de-facto standard for thorough prep. All 150 are in FaangDeck, plus advanced extensions in DP, Graphs, and Greedy that aren't on the list but show up in interviews.
Grind 75
The faster 75-problem subset designed for time-boxed prep (~8 weeks). All Grind 75 problems are in FaangDeck, with the same teaching arc — recognition signals first, then template, then trap, then solution.
Why a superset? The famous lists are great pattern indices but skip the prerequisite scaffolding (Big-O intuition, recursion call stack, hash collisions, the "reading a problem statement" reflex) and the post-mastery extensions (advanced DP, Dijkstra/Bellman-Ford, Union-Find, two-heap median). FaangDeck includes both, so beginners aren't dropped into Two Sum cold and intermediates aren't stranded at Edit Distance.
Famous problems you'll learn: Two Sum, Three Sum, Container With Most Water, Trapping Rain Water, Trapping Rain Water II, Climbing Stairs, Coin Change, House Robber, Longest Common Subsequence, Edit Distance, Word Break, Reverse Linked List, Reverse Linked List II, Linked List Cycle, LRU Cache, Valid Anagram, Group Anagrams, Valid Palindrome, Longest Substring Without Repeating Characters, Longest Substring with At Most K Distinct Characters, Substring with Concatenation of All Words, Min Window Substring, Number of Islands, Clone Graph, Course Schedule, Word Ladder, Reconstruct Itinerary, Bus Routes, Cheapest Flights, Shortest Path Visiting All Nodes, Kth Largest Element, K Closest Points to Origin, Top K Frequent Elements, Merge K Sorted Lists, Find Median from Data Stream, Sliding Window Median, Implement Trie, Word Search II, Subsets, Permutations, Combination Sum, N-Queens, Sudoku Solver, Palindrome Partitioning, Serialize and Deserialize Binary Tree, Construct Binary Tree from Preorder and Inorder, Construct Binary Tree from Inorder and Postorder, Vertical Order Traversal, Spiral Matrix, Rotate Image, Set Matrix Zeroes, Valid Sudoku, Game of Life, Search a 2D Matrix, Insert Interval, Meeting Rooms II, Majority Element, Single Number, Counting Bits, Sum of Two Integers, Valid Parentheses, Min Stack, Daily Temperatures, Best Time to Buy and Sell Stock III/IV/with Cooldown/with Fee, Russian Doll Envelopes, Wildcard Matching, Palindrome Partitioning II, Burst Balloons — plus 150+ more across the 217-lesson catalogue.
Core patterns covered: Hashmap, Two Pointers, Sliding Window, Prefix Sum, Fast/Slow Pointer, Monotonic Stack, Monotonic Deque, Binary Search, Tree DFS, Tree BFS, Trie, Heap / Priority Queue, Backtracking, Graph DFS / BFS, Topological Sort, Union-Find, Dijkstra, Bellman-Ford, Eulerian Path (Hierholzer's), Bitmask BFS, Hashmap + Doubly-Linked List (LRU), 2D Prefix Sum, Matrix Traversal (spiral, transpose, in-place markers), Dynamic Programming (linear, grid, knapsack, string, LIS, interval, tree, state machine), Boyer–Moore Voting, Greedy + Intervals, Bit Manipulation (full-adder, two's complement). The complete DSA interview pattern map.
A six-phase path. ~132 hours. Beginner to interview-ready.
Each phase introduces a small family of related patterns. Phase 1 is the foundation everything builds on; Phase 6 is the optimization toolkit.
Each phase builds on the last. Don't skip Phase 1 or Phase 2 — they're where the patterns you'll re-use everywhere are first introduced.
📐 Phase 1 — Foundations
~6 hours · 11 lessons · do this first
The mental models every later phase assumes: Python you'll use, Big-O time + space, reading a problem statement, the seven-act solving procedure, the core data structures map, hashing internals, recursion and the call stack. An optional Master-Theorem appendix sits next to Big-O.
Covers: Foundations
📦 Phase 2 — Linear DS & Hashing
~15 hours · 27 lessons · build O(n) thinking
The first big family: contiguous data and O(1) lookup. Arrays (Two Sum, Kadane, Container Water, Trapping Rain), HashMaps (Top K Frequent, Subarray Sum K, prefix-sum + hashmap), and Strings (Valid Palindrome, Valid Anagram, Group Anagrams). Every section after this assumes these patterns are reflex.
Covers: Arrays · Hashmaps · Strings
🪟 Phase 3 — Window & Pointer Patterns
~18 hours · 35 lessons · efficient iteration
Techniques for walking linear data structures in O(n) when brute force is O(n²). Two Pointers (3Sum, Sort Colors), Sliding Window (Longest Substring No Repeat, Min Window, Permutation in String), Stacks & Queues (Valid Parens, Min Stack, monotonic stack — Daily Temps, Largest Rectangle), Linked Lists (cycle detection, reverse, merge, fast/slow pointer).
Covers: Two Pointers · Sliding Window · Stacks & Queues · Linked Lists
🌳 Phase 4 — Sorted Search & Tree Structures
~28 hours · 53 lessons · recursion gets real
Binary Search (sorted data + answer-space variants, Koko, rotated arrays), Trees (DFS templates — max depth, diameter, invert, LCA, level order, max path sum, serialize/deserialize, tree reconstruction, vertical order), Tries (prefix tree, autocomplete, Word Search II), Heap / Priority Queue (Kth Largest, Merge K Sorted Lists, Find Median with two heaps, K Closest Points, Sliding Window Median, Trapping Rain II), and Matrix patterns (Spiral, Rotate, Set Zeroes, Sudoku, Game of Life, 2D Prefix Sum, Search 2D Matrix). This is where recursion stops being abstract.
Covers: Binary Search · Trees · Tries · Heap · Matrix
🧭 Phase 5 — State-Space Recursion
~25 hours · 32 lessons · search the space
The recursive search family. Backtracking (Subsets, Permutations, Combination Sum, N-Queens, Sudoku Solver, Palindrome Partitioning — the choose / explore / unchoose template), and Graphs (Number of Islands, Clone Graph, Course Schedule, Union-Find, Word Ladder, Dijkstra, Bellman-Ford, Reconstruct Itinerary, Bus Routes, Bitmask BFS). Identifying when a problem has a tree of possibilities to walk is the key reflex.
Covers: Backtracking · Graphs
🧩 Phase 6 — Optimization & Specialty
~40 hours · 58 lessons · the toolkit
Dynamic Programming Foundations (Climbing Stairs through Knapsack and Word Break), Advanced DP — a full second section covering 2D String DP (Edit Distance, LCS, Distinct Subseq, Regex, Wildcard), Interval DP (Burst Balloons, Palindrome Partitioning II), Tree DP (House Robber III, Binary Tree Cameras), State-Machine DP (Stock III/IV/Cooldown/Fee), LIS family (LIS, Number of LIS, Russian Doll), and counting DP (Combination Sum IV). Greedy & Intervals (Jump Game II, Gas Station, Merge Intervals, Partition Labels, Insert Interval, Meeting Rooms II). Bit Manipulation (Single Number, Counting Bits, Sum of Two Integers, XOR tricks). The optimisation toolkit you reach for when the obvious approach is too slow.
Covers: Dynamic Programming · Advanced DP · Greedy & Intervals · Bit Manipulation
Whether you're starting out or polishing a re-prep.
Full Blind 75 + NeetCode 150 + Grind 75 coverage. Interview Playbook on every problem.
Foundations assumes no CS coursework. Big-O, hashing, recursion all from scratch.
Pattern-first complement to coursework. Sections study independently.
Master decks recover working familiarity with a technique in one short read — your DSA roadmap for re-entering the tech interview market.
Paywalling fundamentals when someone's trying to land their first engineering job is a tax on people who can't afford one. The resources that taught me this stuff were free. I'm paying it forward.
Built as a learning project, not a business. No investors, no growth targets, no surprise upgrade prompts. If it helps, share it with one friend who's prepping.
30 seconds from your first lesson.
No account. No email gate. Pick the start that fits.
Want a note when new lessons land?
Optional. One email when new content drops. No other emails, ever. Unsubscribe anywhere with one click.
What everyone asks first.
The five things you're wondering. Quick answers.
What's the catch? Why is this free?
No catch. FaangDeck is a learning project, not a business — no ads, no upsell, no email-required gate, no surprise upgrade prompts. The resources that helped me learn DSA were free; I'm paying it forward. If you find it useful, share it with one friend prepping for interviews — that's the whole ask.
Is this just another cheatsheet?
No. Each problem is a 24-slide teaching arc — recognition signals, the canonical template, deliberate traps, a full walkthrough, complexity, and an interview playbook. It's a complete lesson, not a summary. If "cheatsheet" sites worked for you, you wouldn't be reading this.
How is this different from NeetCode / LeetCode?
NeetCode is a great pattern index. LeetCode is the world's largest problem bank. FaangDeck is neither — it's a structured curriculum that teaches you to recognize the pattern in seconds. Three signal slides per lesson; same teaching arc every lesson; quizzes + flashcards built in. Covers every problem in Blind 75 / NeetCode 150 / Grind 75 plus extras.
What problems does it cover?
217 lessons across 19 topics, organized into a 6-phase path. Covers the full Blind 75, full NeetCode 150, and full Grind 75 — plus advanced extensions in DP, Graphs, Backtracking, and Bit Manipulation that aren't on those lists but show up in interviews.
Do I need to know Python?
All solutions are in Python, but the visual explanations are language-agnostic — pseudocode-feel, line-commented. If you can read basic code (any language), you'll follow along. Python is the industry standard for whiteboard interviews anyway.
Does it work on mobile? Offline?
Yes to both. It's a PWA — tap "Add to Home Screen" on iOS / "Install app" on Android, and it works offline forever. Adjustable text size, swipe navigation, dark / light / sepia themes, reader mode.
Will there be more content?
Yes. New problem decks, content corrections, and new sections (system design, language-specific tracks, maybe behavioral) drop as they're written. Email above if you want a heads-up.