VisuAlgo

What it is

VisuAlgo was started in 2011 by Associate Professor Steven Halim at the National University of Singapore (NUS) to help his students understand data structures and algorithms through a self-paced, interactive tool. It is still used in NUS courses from introductory programming through to algorithms and competitive programming, and it is free for anyone to use.

Each module lets you build or enter your own input (an array to sort, a set of keys for a tree, a graph) and then watch an operation happen one step at a time, with the matching pseudocode highlighted alongside.

What it covers

Modules include, among others:

• Sorting: the standard comparison sorts and non-comparison sorts, animated bar by bar.
• Linear structures: linked lists, stacks, queues and deques.
• Trees and heaps: binary heaps, binary search trees and AVL trees, and segment trees.
• Hashing: hash tables with different collision resolution strategies.
• Graphs: graph representations, BFS and DFS traversal, shortest paths, minimum spanning trees and more.
• Harder topics such as the travelling salesperson problem, useful for competitive programming.

Features worth knowing

• E-lecture mode: each module has slide-style explanations that walk through the concept before you experiment. NUS uses them for a flipped classroom, where students go through them before class.
• Online quiz: every module has randomly generated questions you can practise in training mode, which is a good check that you can predict the algorithm, not just watch it.
• Your own input: you can run operations on small, hand-picked cases, which is the fastest way to understand edge cases.

Why watching beats reading here

Some structures resist explanation in prose. An AVL tree rotating, a heap sifting an element down or Dijkstra updating distances is a sequence of small movements, and reading pseudocode for it is far harder than watching it happen once. Running it on your own input turns a vague picture into a reliable one.

How to use it well

• When a structure will not click, come here, step through it slowly with a small input, and predict each next move before pressing the button. Being wrong is where the learning happens.
• Try edge cases: insert into an empty tree, delete the root, sort an already sorted array, run BFS on a disconnected graph.
• Take the module quiz after you think you understand it.
• Then write the code yourself. Watching an animation is not the same as implementing it; switch to your editor straight after.

Where it fits

It is a companion, not a course. Use it alongside a data structures course or book (Princeton's Algorithms, a university DSA module or a pattern course) whenever a concept will not form a picture in your head.

Best for: anyone learning data structures, and visual learners preparing for interviews.