caseycrogers/TripleByte Feedback 2018

## TripleByte Feedback 2018
***Personal Story***
Graduated from UC Berkeley 2 years ago. Interviewed with Triplebyte around then, but didn't make it the first time. He joined Google and worked on the firebase team. Now he's looking for something different.
***Communication***
Has a tendency to ramble and sometimes says wrong things confidently. For example, he said that when you send a POST request you also have to specify "the parent path," which is incorrect . This kind of thing happens occasionally.

***Tic tac toe***
He completed 2 steps of tic tac toe. It was slow in large part because he spent a lot of time narrating his process instead of coding. However, he's not a bad coder and the code he puts down is well structured and decent. He made a separate AI class, and seems comfortable with Python.

***Debug***
Very good. His stream of consciousness narration gave insight into his process and he's good at following code and identifying bugs. He fixed 3 and almost fixed the fourth.

***Backend Web***
Backend web wasn't the best. He confused normalisation with data cleaning in a way that suggests that he's not completely sure what a database schema is. The only good answer on the database section was a name drop of "BTree" when we talked about index implementation. Http web wasn't much better, although he explained password salt + hash well.

***Datastructures***
Weak on the details of self balancing trees, struggled with the kth smallest element question, but did ok on everything else on the section.

***Low Level***
Not quite expert-level details, but a very solid section overall.

***Used Car***
Used Car needed some work. He's on the firebase team at google and instinctively suggested firebase for his backend. I probed for details and he struggled a bit. At first he couldn't fully explain his firebase solution beyond "use firebase it'll do everything". Eventually he switched to a relational database backed solution after I asked he how he'd serve a sample query. That pivot didn't improve things much. But I think he just needs to practice design a bit more, familiarise himself with more tools, and practice a bit more on his technical communication.

***Summary***
Despite his weaknesses in back-end web, he seems pretty strong overall. He did well on datastructures and low level and great on debug. He also doesn't seem like a bad coder either. In addition, he has a solid academic pedigree (UC Berkeley) and has worked at Google for the past 2 years. He's enthusiastic and a good culture fit.
	*Personal Story*
	Graduated from UC Berkeley 2 years ago. Interviewed with Triplebyte around then, but didn't make it the first time. He joined Google and worked on the firebase team. Now he's looking for something different.
	*Communication*
	Has a tendency to ramble and sometimes says wrong things confidently. For example, he said that when you send a POST request you also have to specify "the parent path," which is incorrect . This kind of thing happens occasionally.

	*Tic tac toe*
	He completed 2 steps of tic tac toe. It was slow in large part because he spent a lot of time narrating his process instead of coding. However, he's not a bad coder and the code he puts down is well structured and decent. He made a separate AI class, and seems comfortable with Python.

	*Debug*
	Very good. His stream of consciousness narration gave insight into his process and he's good at following code and identifying bugs. He fixed 3 and almost fixed the fourth.

	*Backend Web*
	Backend web wasn't the best. He confused normalisation with data cleaning in a way that suggests that he's not completely sure what a database schema is. The only good answer on the database section was a name drop of "BTree" when we talked about index implementation. Http web wasn't much better, although he explained password salt + hash well.

	*Datastructures*
	Weak on the details of self balancing trees, struggled with the kth smallest element question, but did ok on everything else on the section.

	*Low Level*
	Not quite expert-level details, but a very solid section overall.

	*Used Car*
	Used Car needed some work. He's on the firebase team at google and instinctively suggested firebase for his backend. I probed for details and he struggled a bit. At first he couldn't fully explain his firebase solution beyond "use firebase it'll do everything". Eventually he switched to a relational database backed solution after I asked he how he'd serve a sample query. That pivot didn't improve things much. But I think he just needs to practice design a bit more, familiarise himself with more tools, and practice a bit more on his technical communication.

	*Summary*
	Despite his weaknesses in back-end web, he seems pretty strong overall. He did well on datastructures and low level and great on debug. He also doesn't seem like a bad coder either. In addition, he has a solid academic pedigree (UC Berkeley) and has worked at Google for the past 2 years. He's enthusiastic and a good culture fit.