This article is really difficult to read or its just me? The breaks with the adds are really distracting and I don't know what the author actually is saying here, that he will try windsurf? Do I need to watch the video?
Since LLMs are sometimes wonderfully useful, and sometimes not, I'd suggest effective use involves figuring out in which cases it's likely to succeed, which it's likely to fail.<p>For example, the mentioned graph has "initial prompt with iterative tweaks", followed by iterations of 'starting from scratch'. -- I don't understand why you'd think "this is an ineffective way of doing things", and then keep doing it.<p>Describing LLMs as "slot machines" seems like the author has no curiosity about the shape of what LLMs can/can't do.
I was coding with VSCode for a while and just this week decided to try Cursor which everyone have been raving about. Was quite disappointed to find no meaningful difference with VSCode (so far, after 30 minutes of working with it).
Speaking about agentic coding I have a question about Copilot in VSCode.<p>In agent mode (need to enable it manually in settings), did anyone test forcing Copilot to run unit tests after code changes and fix code to pass tests if they break?
I find VS Code + Cline extension and it's memory-bank pattern far easier to control while still feeling like I'm "vibing". Anyone else use that setup or is it already antiquated?