Coding Practice: Collections

TL;DR

• Map and Set solve a huge number of frontend interview problems cleanly.
• I practice collections as lookup, grouping, counting, dedupe, and cache problems.
• These are often the fastest way to turn nested loops into linear-time solutions.

Context

Collections show up in role maps, route permissions, in-memory caches, deduped analytics, grouped activity streams, and optimistic entity stores.

The Approach

Practice these 10 medium-level questions:

1. Contains Duplicate — classic Set warm-up done in one pass.
2. Two Sum With Index Lookup — Map for complement search.
3. Top K Frequent Items — pair counting with selection.
4. Group Anagrams — normalized keys plus grouped buckets.
5. Build LRU Cache — combine Map ordering with eviction rules.
6. Count Unique Visitors Per Day — nested maps or map-of-sets.
7. First Non-Repeating Character — count then scan, or ordered map strategy.
8. Merge Preference Sources — keyed overwrite with precedence.
9. Track In-Flight Requests — dedupe concurrent promises by request key.
10. Build Reverse Index — map tags to matching record IDs.

Try it in the browser

Trade-offs

• Collections improve lookup speed, but they make iteration order and serialization choices matter.
• Map is often better than plain object for intent clarity, especially when keys are not simple strings.

References

• Related: Coding Practice: JSON & Object Transformations
• Related: Coding Practice: Searching