Day 2: Intervals, Atlassian Interview Questions, and The Diet Problem
Photo by William Liu on Unsplash

Day 2 is in the books, folks!

Yesterday was about binary trees.

Today, I switched gears to Intervals.

Interval problems usually boil down to one core pattern:

Sort by Start Time.

If you don’t sort, you are likely trying to solve it in O(n²), which won’t pass.

The Python sorting check

I had a quick “syntax check” moment today.

When dealing with raw arrays [1,3,2,4], sorting is easy.

But when the input is given as a list of Objects (like class Interval), I always blank out on the Python syntax for a second.

For future reference:

Python

# The clean way to sort objects in Python
intervals.sort(key=lambda x: x.start)

It’s a small thing, but fumbling this in a live coding round is embarrassing.

The Atlassian Variations

I focused on questions that have been tagged for Atlassian.

You know how much they love modifying standard interval problems slightly to throw you off? ;)

1. The Classic: Merge Intervals

Input: [[1,3],[3,5],[8,10]]

Output: [[1,5],[8,10]]

This is standard LeetCode 56.

The logic is simple:

Sort them. If the current interval starts before the previous one ends, merge them.

class Solution:
    def merge(self, intervals: List[List[int]]) -> List[List[int]]:
        intervals.sort()  # Sort by start time
        merged = []
        prev = intervals[0]

        for i in range(1, len(intervals)):
            if intervals[i][0] <= prev[1]:  # Overlap
                prev[1] = max(prev[1], intervals[i][1])  # Merge
            else:
                merged.append(prev)
                prev = intervals[i]

        merged.append(prev)
        return merged

2. The Variation: Ranges with Minimum 2 Overlaps

This was the interesting one.

Input: [[1,3],[3,4],[5,8],[6,10],[9,12]]

Output: [[6,8],[9,10]]

This isn’t just “merge if they touch.”

You need to find the specific timeline segments where the “meeting count” is ≥2

  • Logic: Line Sweep algorithm. You break every interval into points: (start, +1) and (end, -1). Sort the points, then iterate through them maintaining a count. Whenever count >= 2, you are inside a target range.

Meeting Rooms 1/Meeting Rooms II

Solved the classic generic version as well.

It’s essentially the same logic as the problem above, yk, finding the maximum value of concurrent meetings.

The Life Log

  • Activity: 30-minutes walk.
  • Diet: Needs fixing. I’m doing the work on screen, but my nutrition is lagging. This war requires better fuel. I’m open to help here.
  • Random: I shaved today for the first time in an year