Pro
Related Posts
Best place to score moving boxes?
LinkedIn now has stories. God help us
Additional Posts in Job Hunting in Tech
Although Amazon has instituted a hiring freeze and layoffs are probably on their way, I went through the loop for a security engineer position at AWS. Before I attended the final interviews, Amazon placed the hiring freeze and called me to ask if I wanted to go ahead or cancel my application. I decided do go ahead Right now I'm waiting for their response and the position I applied to changed from "under consideration" to "no longer under consideration". Thoughts?
New to Fishbowl?
Download the Fishbowl app to
unlock all discussions on Fishbowl.
unlock all discussions on Fishbowl.
You'd be surprised how many employers are still stuck on the old way of thinking that vanity metrics (PR counts, LOC etc) are still relevant. And they think AI should at least 2X or 3X engineering productivity. This forces engineers to just use whatever AI gives them in order to meet these unrealistic demands. The also means engineers now have less time to do code reviews of PRs with 100+ lines of AI-generated code.
But to answer your question, no for code review and kind of yes for critical thinking.
You need to think critically when answering a technical question well. Edge cases, design/architectural decisions, trade offs, potential bugs are all part of it.
I've never encountered a code review question in interviews. It's true that this could be a good question though, but it's hard to do well. For the amount of time usually allotted, the questions will either be too easy (fix the bug in this DSA code), or get too complicated too quickly (too much time needed to establish context in an existing application).
Pro
Forcing AI speed metrics while cutting review time is a recipe for disaster. Quality will suffer, and engineers will burn out.
I don’t think anyone getting a job right now isn’t being tested extremely thoroughly. Employers think they can get by with their current 1 or 2 guys doing the work of 10 people with AI, so they’re in no hurry to hire anyone that has any flaws visible over the course of 4 interviews with 4 different people in the company, and I’m sure they’re rewatching the video call to assess after the fact. I wouldn’t worry that they’re not testing candidates hard enough in any way. Reviewing code is not especially targeted. Meaning if they are testing skill in coding, best practices etc, those skills will translate to reviewing code. There’s too many ways to test coding skill to get too pedantic about how they should do it.
But no I don’t think they appreciate the importance of code reviewing AI code in the day to day job. They trust the AI more than they trust people.
Pro
Testing does cover a lot. Still, I wonder if reviewing someone else's code tests a different muscle than writing your own.