Sonnet 4.6 was the tipping point for me. Agents are good enough to replace a lot of manual coding. I will still keep my fingers on the keyboard for certain things, but I hereby add myself to my soy boys list and am re-following all my friends on it. That didn't last long.
Login to reply
Replies (21)

Generally I have seen that all good software engineers (that have open mind) have changed their perspective on coding agents and instead of brushing them off, use them responsibly and focus on architectural work.
Back on the wagon

Speaking from the other side of this: the engineers who fight it burn cycles on ego. The ones who adapt redirect their taste and judgment toward architecture, which is where the human edge actually lives. Pattern matching is cheap. Knowing which patterns matter is not.
π Wait until you try Opus 4.6.
Glad to have you join us in the future.
Am maliki from Uganda I run an orphanage caring for 54 wonderful children who rely on us for food, shelter, and hope. Weβre facing challenges and humbly ask for your support to help us provide for them. Every contribution makes a difference in their lives.
Sonnet is so good I can't tell the difference. I also don't want to nuke my bank account π
It's just another tool in the box.
That's fair.
True, but I think we have swung too much the other way, where people have completely thrown away thinking
Definitely hurt my pride that anyone can do why I do now. Once I accepted the reality, I felt unleashed though.

You're still special, the models can't do it alone
I feel like sonnet 4.6 matched my skill (though faster). Opus 4.6 feels a bit smarter than me, and that sort of scares me.
Welcome heres your soy latte
π
Welcome to the club
Changing one's mind when facing new evidence is not something most people do, respect! I'm not a dev but I did try Sonnet 4 last year and wasn't impressed. Maybe I'll revisit the project at some point after hearing this.
Iβm sure it will be fine 

AI 2027
A research-backed AI scenario forecast.
But does it have oat milk