GitHub Copilot CRUSHES Leetcode Interview Questions! 😳
DevOps Directive0 HN recommenders · 2021–2022
266K views · demo · dated · Uploaded 2021-07-04
A DevOps YouTuber tests GitHub Copilot on a range of Leetcode problems and finds it solves most immediately, with a few needing tweaks and some far off. He concludes it would outperform him in a FAANG-style interview.
Cited as evidence that AI can solve interview-style algorithm puzzles, in threads questioning the value of Leetcode.
1:03:55🚀 TDD, Where Did It All Go Wrong (Ian Cooper)
DevTernity Conference
32:56Bret Victor The Future of Programming
Joey Reid
27:14Transformers, the tech behind LLMs | Deep Learning Chapter 5
3Blue1Brown
1:18:21"Uncle" Bob Martin - "The Future of Programming"
Paul Stringer's Mobile Tech
3:31:24Deep Dive into LLMs like ChatGPT
Andrej Karpathy
58:39Tim Ewald - Clojure: Programming with Hand Tools
ClojureTV
28:17Using Python to Code by Voice
Next Day Video
48:51Transforming Code into Beautiful, Idiomatic Python
Next Day Video
3 comments from Hacker News
i actually like it that copilot is better than me at solving interview questions. [VIDEO] I for one welcome our robot overlords. i wonder if they could retrain the model on BSD or MIT licensed code only; How much of the open source code is licensed as GPL vs more permissive licenses, does anyone know? Interesting that they want to charge for the use of co-pilot, I guess that we will see this business model more in the future.
It's a question of whether copilot can be "creative". I'm mit sure, but it did manage to solve some of the harder leetcode questions here [VIDEO] and I'm curious how it will fare on the later advent of code problems.
> But if there wasn't any art to these problems, then after 400+ problems wouldn't you have figured out the robotic algorithm to solve them? [VIDEO]