![[Dev Weekly #112] Andrej Karpathy Joins Anthropic to Lead Frontier AI | The Real Cost of Agentic Engineering | When Rails Breaks at Scale](https://i0.wp.com/d604h6pkko9r0.cloudfront.net/wp-content/uploads/2025/11/24095331/codeminer42_dev_weekly_1920x1080px.webp?resize=1200%2C675&ssl=1)
My First Two Weeks as an Agentic Engineer
by
There’s a version of this post where I tell you AI made me 10x more productive, everything clicked on the first try, and I’m now shipping software while sipping coffee with my feet up. This is not that post. A few weeks ago, Codeminer42 challenged me to build a side project from scratch using only…
AI TIP: Plan Mode into PRD/ERD as Session Handoffs
Plan Heavy with Claude
by
Plan Heavy with Claude: Plan Mode, DDD, and PRD/ERD as Session Handoffs If you’ve used Claude Code on anything bigger than a weekend project, you’ve hit this wall: you’re three hours into a feature, specs are passing, and the session ends, usage limit and context compacted into something unrecognizable. You open a fresh session and…
How to configure Claude Code for any language
A short guide to giving Claude Code the tooling it needs (LSPs, MCPs, language skills, auto mode) so it can self-correct.
by
I mostly work with TypeScript and GDScript, but lately Claude Code has been pulling towards new horizons with Elixir and Rust projects, which I’d never touched before. This post is about what I had to bolt on to make it work. The easy case: TypeScript If you use Claude Code in a TypeScript project, you’ve…
Privacy on Rails
Pragmatically complying with data protection laws
by
On a calm and peaceful day, you’re working on your project, combobulating, discombobulating… then this email shows up in the contact inbox: From: notthatdhh@example.com To: contact@yourcompany.com Cc: veradikt@somelawfirm.com Subject: I’m outta here Hello, here’s DHH (not the one you’re thinking, I’m Dustin Harper Holloway), and my lawyer Vera Dikt who’s cc’d in this email. Send…
Kamal and 1Password: A Safer Deploy Secret Flow
Safer deploy-time secrets for any containerized app
by
The best part of Kamal is that it makes deployment feel smaller. Less ceremony, less platform overhead and less "let’s book infra time next week.” That is why secret handling deserves extra attention. When deploys get easier, it becomes tempting to keep a handful of values floating around in shell sessions, CI settings, or copied…
Opus 4.7 vs GLM 5.1: is mixing models worth it?
by
A couple of months ago, I compared Opus vs GLM by having both of them do a task for me. It’s not that surprising that Opus was best. But what if we get both cooperating? For optimization purposes: one planning, the other executing. I ran this comparison before, same plugin (cm-multilingual), three runs: Opus 4.7…
Brainstorming – The skill that changed Claude for me
How I plan projects using Claude Code and the plugin Superpowers
by
Last Updated on April 29, 2026 How brainstorming improved the way I work with Claude Code, and why you should try it The solo-dev project challenge Every developer has had a project idea, that famous "what if it works?", which received 1 or 2 initial commits and was then archived due to a lack of…
Stop Reading AI Code. Start Measuring It. (A Rails Playbook.)
by
So I was scrolling through Twitter the other day — half-reading, not really looking for anything — and this tweet from Uncle Bob caught my attention: My first reaction was: bullshit, you still have to read the code. Then I actually sat with it for a minute, and I realized I was lying to myself.…
Jujutsu (JJ): An Alternative to Git?
Jujutsu proposes to be simpler yet more powerful than Git. But is it worth switching? Let’s find out.
by
We’ve already spent a lot of time using Git… But another tool is trying to offer enough advantages to justify making the switch. Let’s dig deeper. Table of Contents But how did Jujutsu begin? But why give it a try? JJ vs Git JJ & VSCode JJ in practice JJ useful commands and workflow Next…
Your AI Skills Setup Is Probably Wrong
by
I ran 51 multi-turn evals across 4 configurations and read Claude Code’s source. Skills and CLAUDE.md are both just prompt. The difference is whether the prompt reaches the model.
Codeminer42’s Engineering Blog
