DSA MASTER MAP
SHEETS Two Pointers Sliding Window Binary Search Stack Linked List Sorting Heap Intervals Recursion Trees Trie Backtracking Graph DP Bit Manip Design
130-SECOND TRIAGE // ask these 3 on EVERY problem — they pin ~90% of approaches before you write a line

① INPUT SHAPE

What am I given?

The structure suggests the toolset. Sorted? almost always means two-pointers or binary search.

→ look it up in the Shape Map (§3)

② CONSTRAINT n

How big is n?

The most reliable tell. n fixes your target complexity, which fixes the technique class.

→ read it off the Oracle (§2)

③ WORDING

What phrase does the problem use?

"Number of ways", "top k", "next greater" — exact phrasing is a near-deterministic tell.

→ scan the Keyword Map (§4)
2CONSTRAINT → COMPLEXITY ORACLE // n tells you the target O() — pick the biggest class that still fits
n (input size)target timethat means…
n ≤ 10–12O(n!)permutations · full brute backtracking
n ≤ 18–22O(2ⁿ)subsets · bitmask DP · meet-in-the-middle
n ≤ 100–500O(n³)Floyd–Warshall · interval DP · triple loop
n ≤ 2,000–5,000O(n²)grid / 2-D DP · all-pairs nested loops
n ≤ 10⁵O(n log n)sort · heap · binary search · divide & conquer · segment tree
n ≤ 10⁶O(n) / O(n log n)sliding window · two pointers · prefix sum · monotonic stack · hashing
n ≥ 10⁷–10⁸O(n)single tight linear pass · counting · no per-element log
value ≤ 10⁹–10¹⁸ (not array size)O(log V)binary search on the answer · math · bit tricks
Rule of thumb: ≈ 1e8 simple ops/sec in the room → compute n^? and compare to 1e8.  ·  Space: O(n) aux is usually free; watch recursion depth (stack overflow at ~1e5 deep) and any "in-place / O(1) space" follow-up.
3SHAPE → TECHNIQUE // click a technique to highlight its pattern card (§5)
INPUT — what you're given
OUTPUT — what you must return
4KEYWORD → TECHNIQUE // click a row to highlight its pattern card (§5) · these phrases are near-deterministic tells
techniquesays…
5PATTERN MAP // click a row for the code skeleton · ↳ depth links to the topic's full sheet
6RITUAL + OPTIMIZATION LEVERS // the loop you run out loud, and how to break the brute force
🔧 WHERE'S THE REPEATED WORK? → LEVER
7STUCK? ESCALATION LADDER // run top to bottom — never sit silent
    SHEETS → two-pointers · sliding-window · binary-search · stack · linked-list · sorting · heap · intervals · recursion · trees · trie · backtracking · graph · dp · bit-manip · design
    Every pattern card above deep-links to its sheet · paste dsa-strong-hire-prompt.md into an LLM for a full per-problem walkthrough.