Medium

MediumMatthew Brown – Medium

Matthew Brown

Pinned

Matthew Brown

·

Oct 30

Every Toolkit I’ve Built for Developers Working With AI Coding Agents (All $5)

One page with each problem, the article that explains it, and the kit that fixes it. Start with the one that hurts this week.

Every Toolkit I’ve Built for Developers Working With AI Coding Agents (All $5)

Matthew Brown

·

4d ago

“Is It Nerfed, or Is It Me?” Build a Canary Suite for Your Coding Agent in an Afternoon

Stop arguing from vibes. Twenty tasks from your own repo, pinned settings, and one significance test will tell you whether your model…

“Is It Nerfed, or Is It Me?” Build a Canary Suite for Your Coding Agent in an Afternoon

Matthew Brown

·

Nov 2

November 10 Is a Paperwork Deadline If You Run .NET on Regulated Systems

When .NET 8 and 9 go out of support, your scanners start asking questions. Have the answers ready by Tuesday.

Matthew Brown

·

Oct 30

Your AI Coding Policy Is Either a Ban or a Vibe. Write One People Can Follow.

Eight sections, one CI check, and a tool matrix that covers everything from public repos to CUI, so your developers stop guessing.

Your AI Coding Policy Is Either a Ban or a Vibe. Write One People Can Follow.

Matthew Brown

·

Oct 30

Your USAJobs Resume Is Six Pages Long.

A federal resume is built to pass an HR qualifications check. A corporate resume has to survive a 6-second skim. Here’s the conversion…

Your USAJobs Resume Is Six Pages Long.

In

Let’s Code Future

by

Matthew Brown

·

Oct 30

.NET 8 Goes Out of Support on November 10. Here’s a One-Day Triage if You Haven’t Started.

You won’t upgrade every app in 11 days. You can know which ones matter, what will break, and what to tell your security team.

.NET 8 Goes Out of Support on November 10. Here’s a One-Day Triage if You Haven’t Started.

Matthew Brown

·

Oct 26

Your Agents Write Code Faster Than Your CI Can Tell the Truth

Flaky tests were an annoyance at 5 PRs a day. At 30, they’re the reason nobody trusts a red build. Here’s a triage system for xUnit…

Your Agents Write Code Faster Than Your CI Can Tell the Truth

Matthew Brown

·

Oct 19

Your AI’s Tests Agree With Its Bugs

Green CI doesn’t mean correct code. Here’s how to catch tautological AI-written tests in .NET and Vue projects, with real commands I ran…

Your AI’s Tests Agree With Its Bugs

Matthew Brown

·

Oct 12

Your Coding Agent Can Read Your .env. Here’s How to Actually Stop It

Instructions in CLAUDE.md are a suggestion, not a control. A layered setup for Claude Code, Codex and Cursor, with .NET and Nuxt examples.

Your Coding Agent Can Read Your .env. Here’s How to Actually Stop It

Matthew Brown

·

Oct 9

Your Teammate Can’t Explain Their AI-Written PR. Here’s the Review System That Fixes It.

Five local checks (a size gate, an intent template, an explain-back, a red-before-green test and a risk-first reading order) that make AI…

Your Teammate Can’t Explain Their AI-Written PR. Here’s the Review System That Fixes It.

Matthew Brown

Matthew Brown

I build government software by day and test AI coding tools by night. Writing about .NET, Claude Code and getting real value out of AI agents.

Following

Help

Status

About

Careers

Press

Blog

Store

Privacy

Rules

Terms

Text to speech