Scheitern Coding-Agenten wirklich am Modell?

Thorsten Rinne zeigt, wie Feedback-Loops und fehlender Kontext Agenten ausbremsen – und wie du ein TypeScript-Monorepo agententauglich machst.

👉 https://javascript-days.de/javascript/modernes-js-tooling-fuer-agentic-coding/

#AgenticCoding #CodingAgents

Growing up in the 80s means I like extremely hard to read fonts.

It gets me a lot of complaints from people, so now I just give people the option to change the entire brand of my sites, becasue why not?

#agenticcoding #claude #webdev #ux

Instead of #Python, #JS etc, think #Gleam #OCaml etc and even #Lean4 without using proving capabilities (think ‘partial’) for #agenticcoding loops. Discrepancies in training data size can be alleviated by creating synthetic code. These langs give top feedback at compile time. #AI labs listen
3/3

My coding agents can finish six features before I’ve reviewed the first.

Now each merge request comes with a short before-and-after video. I see what changed, then dig into the code.

Here’s the workflow:
heristop.github.io/blog/2026-09... #AgenticCoding

Claude Fable 5.1 is built for long coding, research, and knowledge-work projects—without constant back-and-forth. #ClaudeFable #ClaudeAI #AgenticCoding #AIResearch #KnowledgeWork

IMO: hyperscaling is not the solution*) in the long run. You can run open competitive enough models in your own PC. If not now, in very near future yo will be.

*) For governmental/national security tasks, for sure, they are. But not for e.g. coding.

#AI #LLM #Cybersecurity #AgenticCoding