margin

Since you were last here

I found a few things worth talking about.

4 drafts are ready for you, 2 ideas need one detail from you, and I spotted 7 new angles in what you shared this week.

What's on your mind?

Messy is fine. I'll tell you what I understood first. ⌘↵

Drafts for you

Written from what you shared. Nothing is posted without you.

Ready to reviewUpdated

AI coding gets interesting before the coding starts

Recently I built a plugin that can take a task number, fetch the ticket, check the repository, and plan the work before writing a single line of code. I built it for Claude Code because I was tired of the same routine. Open the ticket. Read it twice. Search the codebase for the files it probably touches. Work out what's already there. Only then start. What surprised me is that the most useful part isn't the code it writes. It's the plan. When the AI has read the ticket and the repo first, the conversation changes. It stops guessing. It asks better questions. It points at the files that matter instead of inventing new ones. I think a lot of us use AI coding tools backwards. We ask them to write first and understand later. The interesting part of AI coding happens before the coding starts. Do you let AI plan before it writes, or does it go straight to code for you?

You said the planning part is more useful than the coding part. I built the post around that line instead of the feature list.

Based on: Claude Code plugin ·
Review draft
Draft improvedUpdated

Stop asking AI to write code first

I think developers are using AI completely wrong. We open a chat and ask it to write code. A component, a hook, a migration. Then we spend the next twenty minutes fixing code that doesn't fit the project. The real leverage, at least for me, is asking it to read first. Before I let it write anything now, I ask three things: — Which files would this change touch? — What patterns does this codebase already use for this? — What could break? The answers are usually more useful than the code. AI is a decent writer of code. It's a much better reader of it.

Based on: AI coding opinion ·
Review draft
New angle foundUpdated

Agents won't remove the boring work. They'll expose it.

Everyone is reading the new agent announcements as “AI will do the boring parts of engineering now.” I think it's the other way around. Agents won't remove the boring work. They'll expose it. An agent that opens tickets, runs tests and updates statuses needs all of those things to be clear. Vague tickets, flaky tests, deploy steps that live in one person's head — humans quietly work around these every day. An agent just gets stuck. I've seen a small version of this with my own Claude Code plugin. The places where it struggled were exactly the places where our process was fuzzy. So before asking “what can agents do for us?”, it's worth asking “what would an agent trip over in our workflow?” That list is probably your real backlog.

Based on: OpenAI's new agents for developer tools ·
Review draft

Post opportunities

Ideas I noticed in what you shared. Develop one and I'll start a draft.

From your voice note ·

“Something interesting happened during our architecture discussion today. We spent like forty minutes debating whether to split the dashboard into micro-frontends. Everyone had opinions about module federation and build times. Then someone asked, okay, who is actually going to own each piece? And nobody could answer. We ended up keeping the monolith for now. I think the whole debate was about the wrong thing.”

  1. 1.

    Architecture decisions are really ownership decisions

    Angle: Engineering thinking

    Develop
  2. 2.

    We almost split our frontend for the wrong reason

    Angle: Story / lesson

    Develop

From your thought ·

“Recently I built a Claude Code plugin that can take a task number, fetch the ticket, check the repository to see what's already there, and plan the work before writing anything. It also updates the task status when it's done and can point out blockers before implementation starts. Honestly the planning part is more useful than the coding part.”

  1. 1.

    I stopped manually preparing every development ticket

    Angle: Workflow automation

    Develop
  2. 2.

    The best thing my AI plugin does isn't writing code

    Angle: Contrarian / insight

    Develop

From your screenshot ·

“A product manager asks whether we can “just add a region field” to the Global Operations Hub screen. Replies from three different teams show each of them uses that screen for something different — one for reporting, one for daily dispatch, one for audits.”

  1. 1.

    “Just add a field” is never just a field

    Angle: Product engineering

    Develop

Saved for later

Ideas you kept. I'll remember them until you're ready.

  1. I explain things better when I draw them

    Angle: Mentoring

    From your thought: Pairing with a junior ·
    Develop