It's manic at first, but I have become a big fan of developing with #roborev
It's manic at first, but I have become a big fan of developing with #roborev
Fancy things happening here: #Qwopus
Jackrong on hugginface tried teaching the qwen3.5 models the thinking process of Anthropics Opus 4.6 [1]
Can‘t wait trying that out 🙂
https://huggingface.co/collections/Jackrong/qwen35-claude-46-opus-reasoning-distilled
Wer lange keinen Code geschrieben hat, baut besser keine Software mehr. Eigentlich. Wie aus einer guten Idee lange nichts und mit Hilfe von KI-gestütztem Entwickeln doch noch etwas wurde (ein Selbstversuch): https://coma.photography/blog/2026-03-14-relaunch/
I built GitTop this weekend, a terminal dashboard for visualizing Git repo stats, inspired by htop/btop. It visualizes commit heatmaps, contributor profiles, branch stats, file health, and more.
Incidentally, it is also my first fully agentic coding project.
The result is good: braille area charts, fractional Unicode bar rendering, a proper filter DSL with boolean logic.
If you're a developer who feels like you're struggling to catch the immense AI wave that has stormed into our lives, I can identify! For over a year, I like many others had been using AI to produce code in a way that constantly required my intervention. When I started to give agents the tools that I had myself access to, it really did start to shift the needle. So much so that I'm embarking on a blog series on "Seven Wonders of Agentic Coding", which will cover the tools I've given to the agent, and the results I've seen.
Cursor (@cursor_ai)
에이전트형(agentic) 코딩 과제에서 모델을 평가하는 새로운 점수화 방법을 공개했습니다. 이 방법으로 Cursor 내 여러 모델의 지능(intelligence)과 효율성(efficiency)을 비교한 결과를 공유한다고 알리며, 코딩 에이전트 성능 평가에 대한 새로운 벤치마크 또는 메트릭 제안을 포함합니다.
I’ve become a manager of AI agents. I have avoided managing people for most of my career. Now I get to manage machines. Jury is out on whether it is easier.