A phone interview about my general work history led to an in-person meeting. The question asked was: "In a given string, find the last unique character." My initial approach was to filter out special cases, such as an empty string, a null string, a blank string, or a single-character string. Some of this required firming up requirements, for example, "What do you return for a null string?" The interviewer's eyes rolled when I asked this, and I was told not to worry about it and just code the answer.

The actual coding was to be done on a whiteboard. My solution was to start from the end of the string and loop backward. For each character, I compared its first index position against its last index position. If the first and last index positions were the same, the character was unique. Since I started from the end, the first time this occurred was the last unique character. When that happened, I would exit the loop and return the answer.

I was told this was incorrect. The interviewer explained the correct way was to loop through the string, hash the characters, and keep track of the current last hash created. When the loop finished, the last hash created would be the last unique character. We then discussed the Big O values of each method.

Overall, the interview was difficult. The interviewer gave the impression their time was being wasted right from the start.

Find the last unique character in a character string.