Skip to content

About

Hi, I'm Ethan.

I write about how software teams actually use AI in their work. Not the keynote version. The Monday-morning version.

The site has 4 lanes: SOPs and playbooks you can lift into your own team, case studies of what worked and what didn't, field notes on adoption, and talks I've given on the topic.

If you want to compare notes, book a workshop, or invite me to speak, the contact form is the fastest way in.

What I write about

AI SOPs in delivery

Most teams treat AI as a personal productivity hack. The real payoff shows up when it becomes a team SOP: how we review code, how we write tickets, how we run standups, how we test. I document the ones that actually hold up.

AI in the real workflow

I'm not writing about prompts in isolation. I'm writing about how prompts fit into the moment between "I have a Jira ticket" and "I've merged the PR." That's where most of the value and most of the friction lives.

At all scale

What works for a solo dev breaks at 10 people. What works at 10 breaks at 100. I write scale-aware SOPs so you can pick the version that fits your org today, and the version you're growing into.

AI adoption

Tools are easy. Adoption is hard. I write about the change-management side: pilots, measurement, holdouts, training, and what to do when a team rejects something.

How I work with AI

I use AI inside every step of my own delivery work: planning, writing, coding, reviewing, communicating. Claude is the daily driver, with model and tool choices that change as the work changes.

My rule is simple. AI in the workflow, not on the slide. If a tool doesn't move actual delivery faster or cleaner, it doesn't earn a spot in my stack.

Before I write a SOP or a playbook, I run it on real work for a few weeks. If it survives, it ships here. If it doesn't, it goes in the field notes with what broke.

What I believe

  • •AI is a workflow question, not a feature.
  • •Reliability beats novelty.
  • •SOPs beat vibes.
  • •AI-fluent teams out-deliver bigger teams that aren't.
  • •Adoption is a measurement and habit problem, not a tool problem.
  • •Honest field reports beat hot takes.

Workshops, talks, or advisory?

About — Ethan (Duy) Bui