Every year, Ruby Central's Scholars & Guides Program pairs aspiring Rubyists with experienced mentors for a week of learning, connection, and growth at RubyConf. Scholars get a Guide to help them navigate the conference, build their coding skills, and meet other Rubyists face to face. Alongside that mentorship, Scholars are also invited to take on a mini-project, an open opportunity (not a requirement of the program) to share their own unicorn, whatever makes them uniquely them, with a community that's wonderfully diverse in thought, interests, skills, and talents. This year's theme was "Ruby for funsies: using Ruby and Rails to enhance a hobby," and Scholars could present however felt right to them, whether that meant a blog post, a webpage, a video, or a Lightning Talk on stage.
We want to spotlight one Scholar's project in particular, because we think it says something important about where our community is headed.
Meet Madeline Caples
Madeline Caples is a frontend developer who came to RubyConf 2026 as a Scholar and left with a new project, a Notes app built in Ruby on Rails, and a lot to say about how they got there. For their mini-project, they wrote a thoughtful blog post called "Learning at Ruby Conf: A Workflow For AI Assisted Study," which tackles a question a lot of developers are quietly wrestling with right now: how do we keep learning and growing as programmers when AI can write working code from a single prompt?
Madeline has written about this tension before. Earlier this year, they published a piece explaining why they were hesitant to lean on AI while learning, worried it would shortcut the hard-earned understanding that makes someone a confident developer. RubyConf gave them a reason to revisit that stance, not to abandon it, but to refine it.
A workflow built for real understanding
Working on their Notes app during the conference, Madeline developed a five-step study flow they now use whenever they sit down to learn something new with AI tools like Claude and Cursor by their side:
- Define a clear goal and a realistic time frame before starting
- Attempt the problem on their own first, taking notes and connecting new ideas to what they already know
- Correct their work with AI feedback, but pause to understand why something was wrong
- Refine their understanding by exploring edge cases
- Iterate, repeating the cycle for each new goal
The result is a simple, structured way to use AI as a study partner rather than a shortcut, one that keeps the learner doing the actual learning. It's a generous, honest piece of writing, and it's worth reading in full.
You can find Madeline's full write-up on their blog, <div>ersions</div>, where they regularly write about machine learning and software engineering topics in plain, accessible language.
Why this program matters
Madeline's project is exactly what the Scholars & Guides Program is meant to make possible: a space for someone to bring a real question they're wrestling with, spend a week surrounded by a community that takes that question seriously, and walk away with something worth sharing. That's true whether a Scholar's mini-project takes the form of a blog post, a Lightning Talk, or something else entirely. There's no requirement to get on stage. The point is simply to share who you are and what you're working through with a community that wants to hear it.
If you're curious about applying as a Scholar, or becoming a Guide yourself, keep an eye on Ruby Central's announcements for the next cycle. And in the meantime, go give Madeline's post a read. It might just change how you think about your next study session too.