Steven Li
@realstevenli
Prev engineer @coinbase. Built startups @routefireio, @rooferdotcom. Also running a group chat for founders with $100k+ ARR, DM for access.
ID:866834282026684416
https://steven4354.github.io/ 23-05-2017 01:52:52
1,6K Tweets
2,2K Followers
1,3K Following
AI software engineers are compared to humans based on SWE-bench, which mainly covers Python tasks with ≤15 line changes evaluated by unit tests. I wrote this article to provide a framework to assess if AI's progress on this benchmark is relevant to you
stepchange-blog.ghost.io/why-do-ai-soft…