Read My Thinking

I try it myself. Then I share what I learn.

I write about the decisions behind an experiment: what I was trying to do, where the first approach broke down, and what I changed. A finished app or a conference session rarely shows that part.

Teaching myself new tools gives me experiences to examine and share. I include uncertainty and setbacks because they give other people a more realistic starting point for their own work.

Below are stories, lessons, and questions from those attempts, including AI in the RTE role, peer learning, and turning ideas into outcomes. Take what is useful, test it in your own setting, and let me know what you learn.

Latest updates

What I'm working on

Updated October 2026

  • Hosting live peer learning sessions. A small group, one real AI experiment each, and something to try on Monday. The first one was October 21. The next one is Wednesday, December 2, at 11 AM Central. Want in on a future one? Message me on LinkedIn.

    October 2026
  • Spoke at the 2026 SAFe & AI Summit in San Diego. Read the story.

    September 2026
  • Sharing what I’m learning about technology, work, people, and my own responses to change in Alex 2.x: Always the latest version of me on LinkedIn.

    New
  • Launched AIQ RTE, a free assessment for Release Train Engineers. Twenty questions, under five minutes.

    Live

Ideas in practice

Explore by theme

Selected posts from LinkedIn, grouped by theme. Each one opens the full post.

Lessons learned

Sharing the experience I almost kept to myself

“Sometimes the experience we dismiss is exactly what someone else needs to hear.”

I nearly talked myself out of a Summit session because I didn't feel like a thought leader. I told what happened to me instead.

Read on LinkedIn →

Lessons learned

Rebuilding the polling app ten days before the Summit

“We learn AI faster from each other's misses than from anyone's highlight reel.”

Ten days out, I rebuilt the polling app I'd made with AI: one page, one link, refreshing live. On the day, it worked.

Read on LinkedIn →

Peer learning

Bring one experiment. Learn from each other.

“Not another webinar. Not a presentation.”

Small peer conversations where everyone brings one real AI experiment: a win, a miss, or a work in progress.

Read on LinkedIn →

Capacity

Two questions about RTE time and human responsibility

“AI can support the role. It should not own the human parts of it.”

Where RTEs say their time goes, and what they won't hand to AI, were two different lists. That gap was the whole session.

Read on LinkedIn →

SAFe, going AI-Native

Six sessions on SAFe’s shift to AI-Native work

“The hard part is the ways of working.”

The recap of my six-part series on what changes for SAFe, its roles and the people in them.

Read on LinkedIn →

Capacity

The work-around-the-work ate my day.

“What I did with that time mattered more than the time itself.”

Small AI experiments gave me back time as an RTE, and I spent it with my team.

Read on LinkedIn →

Outcomes

From an AI idea to measurable business results

“It's turning AI into measurable outcomes.”

The organizations that pull ahead will get their leaders aligned, change how the work gets done and bring their people with them.

Read on LinkedIn →

Keep learning with me

Alex 2.x: Always the latest version of me.

What I’m learning, trying, and changing, shared on LinkedIn.

Follow along on LinkedIn →