A month ago I finally marked something off my todo list that had been there for a couple years. I went through an interview process as an interviewee.

I did a 2 hour video-call with Triplebyte. And unlike basically every company ever, Triplebyte actually gives feedback to candidates so I know slightly more than just "I passed."

The interview was broken up into sections, and one section in particular was super interesting.

Initially I thought the section was similar to an interview I would give at Khan Academy as part of our Senior Engineer pipeline. During my Khan Academy interview, I'd have the candidate essentially roleplay being a senior engineer at a fictional company who's been tasked with creating some big feature. My interview was trying to go after many skills, but one of the primary skills was the ability to smash at ambiguity at lots of levels. Jumping forward while making tons of assumptions was A Bad Thing™️.

So when Triplebyte game me a high-level prompt for a webapp and asked me how I'd build it, I tried to smash at the ambiguity in the prompt and figure out what they were asking for. But I kept hitting walls with the interviewer. I could tell he wasn't prepared for the questions I was asking, a sign that I'm barking up the wrong tree. We moved on pretty quickly so I'm not sure what sort of conversation he was actually looking for, even in hindsight, but I'm almost certain it was a technical conversation and not a PM-like-ambiguity-crushing one.

My official feedback had the diagnosis that I know nothing about system design. A bit harsh 😅, but I see how they they arrived there since I don't think I proceeded at all along the direction they wanted me to during this section (which was the "System Design" section).

This has got me thinking a lot about the interviews I've given. Not being clear about what I was looking for was something I was always worried about when I was conducting interviews, and seeing it so clearly from the side of the interviewee is super illuminating.

Triplebyte is in the business of interviewing. They're about as close to professional interviewers as you can get, and yet I have no idea what aspect of building that webapp they wanted me to talk about.

And ultimately, they came out with a silly diagnosis because of it. They failed to extract truths about me. Saying my opinions around system design are silly is totally reasonable but there should be no doubt that I know things about it.

And the interviewer sounded very confident in that diagnosis.

How many times did I reject a candidate after my interview, with reasonable confidence that they would have trouble hacking through ambiguity if they joined Khan Academy, while they would've actually been fantabulous at it. My bet is so many times.

(I've got lots more thoughts on interviewing. Follow my dinky blog if you wanna read those thoughts 😃. Also I'm not looking for work right now, I just had a golden opportunity to do an interview and decided to take it.)


« more posts | follow me via twitter, or RSS