Welcome to the DSA with Python track — a complete, hands-on path from “what is Big-O?” all the way to segment trees and max-flow. It is built for two goals that overlap more than people think:

• Cracking big-tech interviews (FAANG-style problem solving).
• Competitive programming (LeetCode, Codeforces, contests).

What you’ll learn

• What data structures and algorithms actually are, and why they decide whether your code runs in 0.1 s or times out.
• How to run and solve problems right here in the browser — no setup.
• A repeatable way to read a problem, spot the pattern, and reach for the right tool.
• The full toolkit: arrays → trees → graphs → dynamic programming → advanced CP.

Why DSA matters

An algorithm is a step-by-step recipe. A data structure is how you organize data so an algorithm can be fast. Pick the wrong structure and even a correct algorithm is too slow to pass.

The same task — “is xx in this collection?” — costs wildly different amounts depending on the structure:

That single choice — listlist vs setset — is the difference between Accepted and Time Limit Exceeded on a large input.

The roadmap

This track is organized into ordered phases. Follow them top-to-bottom, or jump to a phase you need.

  graph TD
    A[Phase 1: Foundations
Big-O, recursion] --> B[Phase 2: Python for DSA & CP
stdlib, fast I/O, TLE]
    B --> C[Phase 3: Core Data Structures
arrays, trees, graphs, heaps]
    C --> D[Phase 4: Sorting & Searching]
    D --> E[Phase 5: Interview Patterns
the ~15 named patterns]
    E --> F[Phase 6: Recursion, Backtracking & DP]
    F --> G[Phase 7: Graphs Advanced]
    G --> H[Phase 8: Advanced CP Topics]
    H --> I[Phase 9: Templates & Cheatsheets]
    I --> J[Phase 10: Interview & Contest Strategy]
    J --> K[Phase 11: Problem Sets
LeetCode-style practice]

How to use this track

Every code block on this site is live. It runs real Python in your browser (via Pyodide) — click Run to execute, edit the code, and re-run. Try it:

# Edit me, then press Run.
nums = [5, 3, 8, 1, 9, 2]
 
# Two ways to find the max — one is O(n), one hides a sort at O(n log n).
print("max via built-in:", max(nums))
print("sorted[-1]     :", sorted(nums)[-1])

# Edit me, then press Run.
nums = [5, 3, 8, 1, 9, 2]
 
# Two ways to find the max — one is O(n), one hides a sort at O(n log n).
print("max via built-in:", max(nums))
print("sorted[-1]     :", sorted(nums)[-1])

You’ll also see three other block types throughout the track:

• p5p5 animation panels — watch an algorithm move (sorting, BFS, DP fills).
• mermaidmermaid diagrams — trees, graphs, and recursion structure.
• Graded exercises — small tasks that check your answer instantly.

A first taste of “which structure?”

Below is a classic interview warm-up: does the array contain a duplicate? The naive double loop is $O(n^2)$. A setset makes it $O(n)$. Fill in the blank to use a set.

Recap

• Data structure = how data is organized; algorithm = what you do with it.
• The right structure turns a slow solution into a fast one — often $O(n)$ vs $O(n^2)$.
• Every code block here is runnable; edit and experiment freely.

Next: Setup for CP & interviews — accounts, tooling, and how to practice.