DSA pattern
Graph Algorithms
BFS explores nodes level-by-level (using a queue), perfect for shortest path in unweighted graphs. DFS explores as deep as possible first (using recursion or stack), ideal for cycle detection, topological sort, and connected components.
When to use
5 signals- 01Shortest path (unweighted) - BFS
- 02Cycle detection - DFS
- 03Topological sort - DFS
- 04Connected components
- 05Grid/island problems
In interviews, call out the pattern early. It helps interviewers follow your reasoning and keeps your implementation focused. After solving, explain tradeoffs in plain language: why this approach is efficient, where it can break, and what edge cases you would test first.
Practice problems
4 listed- 01Number of IslandsPractice →
- 02Clone GraphPractice →
- 03Course SchedulePractice →
- 04Word LadderPractice →
Build a review loop
Treat each pattern as a repeating cycle, not a one-time read. Solve a fresh problem today, revisit the same pattern in 2-3 days, and run one timed attempt at the end of the week. This turns short-term familiarity into reliable recall for live interviews.
When reviewing, re-derive the core steps from scratch before checking a reference solution. If you can explain the state transitions and edge cases clearly, you are much more likely to execute well under interview pressure.