1st Coding: Something related to interval. Saw this for the first time. In hindsight, question was failrly easy but required some thinking to solve it optimally. But, at that point of time, couldn't think fast. Was able to solve optimally at the end. 

2nd Behavioral: Standard questions - conflict, best project so far etc. Coding Q later.

3rd System design: Twitter keyword search

4th Coding: Graph related question - finding path from top left corner to bottom right corner in a table full of 1's and 0's where 0 means you can go through while you can't if its 1. (please point out if you've seen this on Leetcode)
Solved it using dfs. Interviewer asked if you could solve it using bfs as well and if so, whats the benefit. I went through an exmaple and mentioned it seems both the approaches will have same result. He gave me an another example where I could see how bfs would give a shorter path. At last, he asked me to tell my approach using bfs. (please point out how you would use bfs here along with complexities for both the dfs and bfs). Also, asked me to analyse complexities of both the appoach. I told him O(n*n) as in the worst case, we would end up visiting every cell for both the approaches.