Hosted by Jackrabbit Learning Experience (JLX) in October 2025, the AI Forum for Learning Professionals is a series of webinars that brings together educators, instructional designers, technologists, and learning leaders to explore how artificial intelligence is reshaping the way we design, deliver, and imagine learning.
Visit our AI Forum page to recap other brilliant sessions.
If you wish to receive updates about our next AI forum, please subscribe to our newsletter now.
Session 3: Hands-on Claude Code Workshop by Brick Maier
March 5th, 2026
This week learning professionals gathered once again for another AI Forum for Learning Professionals, and this one had a different energy. It was hands-on, live, and at times delightfully unscripted. Instead of talking about AI tools in the abstract, we watched someone use Claude Code in real time, narrating their thinking, hitting unexpected moments, and improvising on the fly.
The result was one of the most practical sessions we’ve had yet.
If you missed our previous forum recaps, you can find them below.
Gleeful and curiosity-driven learning professionals collaborating for the common cause of learning and community support.
The Presenter: Brick Maier
Our guide for the session was Brick Maier, Director of Enterprise Learning at a company focused on workforce development in the renewable energy sector. Brick has spent the past year using Claude Code daily, not as an experiment, but as a genuine part of how he works. He came not to sell the tool, but to show it honestly, in action, with all the texture of a real workflow.
The room felt that immediately. As one attendee put it afterward:
“Really helpful for a complete novice to this. Made it digestible and less intimidating.”
The Big Idea: Claude Code as Orchestrator, Not Just Assistant
One of the first things that landed with the group was a reframe around what Claude Code actually does.
Brick described it this way: when you’re using Claude Code, you’re not writing code line by line. You’re more like a conductor. You give direction, provide context, set guardrails, and make the calls about when to push forward and when to stop and verify. Claude Code handles execution while you handle intent.
This sparked immediate recognition in the room. Several attendees noted that this mirrors what’s happening across learning design more broadly, the shift from producing to directing, from making to deciding what gets made and why.
Brick also introduced a concept that resonated strongly with the group: the cost of rework. When Claude Code is building things on your behalf, the quality of what you get out is directly tied to the quality of what you put in, your research, your planning documents, your contextual guidance. A little upfront investment in thinking saves a lot of downstream fixing.
What the Live Demos Showed
Three live demonstrations walked the group through different moments in Brick’s real workflow.
Explore the presentation slides and demos here.
Demo 1: Adding a feature to an existing tool
Brick showed how a working application, something built for his own use, could be extended with a new capability through conversation with Claude Code alone. No IDE open, no dragging files across panels, just describing what was needed, reviewing what came back, and deciding whether to push further or save the work.
What struck people was that Brick wasn’t reviewing every line of code. He was checking outputs, validating behavior, and making judgment calls. The act of getting commit, saving a checkpoint before moving on, was framed as a discipline rather than a technicality. You’re not trusting Claude Code to remember. You’re trusting yourself to save.
Demo 2: Querying a live database through conversation
This one generated audible reactions. Brick connected a real data source to Claude Code and simply asked questions about it in plain language. Claude Code wrote the queries, pulled the results, and surfaced patterns.
The point wasn’t the specific data. It was the shift in what becomes possible. Tasks that used to require specialized technical skill, or a request to a data team, can now happen in the flow of your own work. Professionals who know their domain deeply but don’t have a coding background are increasingly able to act on that knowledge directly, with Claude Code doing the technical heavy lifting.
Demo 3: Building and deploying a small web application, live
The third demo pushed further. Starting from a planning document and a clear intent, Brick walked through building a small, functional application using Claude Code and deploying it live to the web in under ten minutes, not to a staging environment but to a real URL.
The group watched a GitHub push trigger an automatic deployment, watched the site refresh, and watched Brick make a cosmetic change mid-session through a conversation with Claude Code and see it reflected in real time.
The room got quiet during this one, in the good way.
The Workflow Beneath the Demos
What made the session valuable wasn’t just the moments of “look what it can do.” It was seeing the discipline Brick had built around using Claude Code consistently.
A few principles came through clearly:
Planning documents as input. Before Claude Code writes anything, Brick invests time in research, context-setting, and writing out what he wants. That document becomes Claude Code’s operating instructions, and the quality of the plan shapes everything that follows.
Test-driven thinking. Rather than building in large chunks and hoping it works, Brick described working in small, testable increments, validating each piece before moving to the next. He didn’t frame this as a coding methodology. He framed it as simply not getting lost.
Knowing when to save. Brick was transparent about the experience of going too far without committing, running out of context, losing the thread, and having to start over. Getting commit became a recurring theme throughout the session: a habit of pausing, validating, and preserving good work before pushing further.
Letting Claude Code work, but staying in the room. There were moments where Claude Code was visibly working, planning, reading files, drafting, and Brick watched it rather than stepping away. Not micromanaging, just staying present so he could course-correct quickly if needed.
What the Group Brought
The Q&A opened up genuinely useful territory, and a few threads are worth naming.
On cost and access. There was a practical conversation about how Claude Code is priced, how context works, and how to get meaningful results without burning through tokens unnecessarily. The short version: smaller, smarter models often outperform heavier ones for focused tasks, and knowing how to work within limits is itself a skill worth developing.
On tone and voice in Claude Code-generated content. One question surfaced something the group spends a lot of time on: how do you tune Claude Code’s output so it doesn’t sound generic? The answer, in practice, involves examples, explicit guidance, and iteration. With the right prompting and context, the gap between default and excellent closes significantly.
On human judgment in the loop. A recurring question across the session was where the human needs to stay firmly in charge. Brick was clear: deployment decisions, quality validation, and anything that touches real users require a person to make the call. Claude Code doesn’t know what “good” looks like in your context. You do.
On the changing shape of roles. This came up more than once. If Claude Code can handle execution, what does that mean for professionals whose value has been tied to execution? The conversation didn’t resolve neatly, but the emerging consensus was that domain expertise, judgment, and the ability to direct well become more valuable, not less. The work shifts. It doesn’t disappear.
What Attendees Said
The feedback after the session said it all.
“Really helpful for a complete novice to this. Made it digestible and less intimidating.”
“First AI webinar I’ve attended in about a year that has actually been super worthwhile. Was engaged the entire time.”
That second one landed especially hard. There’s a lot of noise in this space right now, and sessions that cut through it, leaving people more capable rather than more anxious, are worth paying attention to.
A Closing Thought
Something Brick said near the end has stayed with me.
He described the experience of watching Claude Code take a vague instruction, something said almost offhandedly in the middle of a live session, and just go do it. Correctly. Quickly. Without being prompted again.
His reaction in the moment was something like disbelief, followed by a quiet but important realization: I need to be the one deciding when that happens.
That’s the posture this session kept coming back to. Not awe, not anxiety, just a clear-eyed sense of responsibility. Staying in the role of the person who decides what gets built, what gets deployed, and what it’s for.
That’s still a very human job. And it’s getting more important, not less.