Nathan Wrigley interviews Robby McCullough, co‑founder of Beaver Builder, about the product’s evolution, how the team is approaching AI, shifting site‑building workflows, and what’s next for the WordPress ecosystem.
Background and timeline
Beaver Builder is entering its twelfth to thirteenth year as a WordPress page builder. Robby and his team began as an agency and created the builder so clients could edit their own sites instead of relying on developers for every change. Lately Robby has balanced product work with major personal events — a newborn and a move — while steering the company forward.
A cautious approach to AI
Rather than rushing to bolt “AI” onto features, Beaver Builder waited while early LLM integrations proved to be more window‑dressing than substance. Robby argues that adding lightweight AI—like auto‑generated headings—has limited value; meaningful benefits arrive when the underlying models and tooling can solve problems that previously weren’t worth building around.
Two experimental AI directions
The team is exploring two main ideas. First, what Robby calls “vibe coding”: take an externally generated landing page and import it into Beaver Builder, turning it into a fully editable page in the builder so users can drag, tweak, and publish immediately. Second, an in‑context chat agent that works inside an existing page or site, focusing on a specific region (for example, a pricing table) and helping with copy, layout, or markup adjustments. Both concepts are experimental and being kept internal until they’re ready.
Page builders versus AI site generation
Robby and Nathan agree that page builders are far from obsolete. Builders removed a large amount of friction from site creation, enabling many more people to make sites and likely contributing to WordPress’s adoption. AI that can spit out entire static sites quickly is useful for one‑off brochure pages, but WordPress’s extensibility and administrative plumbing—drafts, images, metadata, plugin ecosystem—still give it a strong edge for projects that need longevity or customization.
Transparency over black boxes
Beaver Builder intends to avoid turning the product into a closed, tokenized black box. They want AI features that expose front‑end code—markup and CSS—so developers and learners can inspect, modify, and improve what the AI creates. Robby envisions users bringing their own AI key or agent, granting it access to Beaver Builder while retaining the ability to see and change the generated code.
Craft, nostalgia, and the role of hands‑on skills
Both hosts reflect on the loss of hands‑on practices—manual HTML/CSS, pixel‑level work in design tools—replaced by faster, more automated workflows. Nathan compares the shift to preferring carpenters over flat‑pack furniture: a tradeoff between artisan skill and speed. Robby still enjoys iterative design: he uses AI to generate ideas quickly, then tunes CSS and layout by hand. He’s found agentic tools helpful to learn modern CSS techniques like grid and variable fonts. They don’t expect craftsmanship to disappear entirely: some people will remain specialists, and many projects will retain traditional workflows.
Maintenance and the lifecycle of sites
Nathan points out an often‑ignored issue: maintenance. Creating a site quickly is one thing; keeping it running and relevant over years is another. Page builders could become the editing and maintenance layer for sites initially produced by AI, serving more as ongoing managers than as mere generators. Robby agrees, noting WordPress’s strengths for long‑term customization and business‑critical sites.
Business planning amid fast change
They discuss the uncertainty product teams feel as AI capabilities accelerate. Robby describes himself as a hopeful optimist but acknowledges that the industry has faced existential scares before—Dreamweaver’s reputation, Gutenberg’s introduction, predictions of page builders’ death—and yet adapted. Both recognise that rapid AI progress could still force significant shifts in how products are designed and supported.
Practical AI use and personal stories
Robby shares everyday examples of how AI helps: using dictation when caring for his infant to capture ideas, and a side project using ChatGPT to scrape local ham radio repeater info and produce formatted CSVs. These anecdotes illustrate how AI can eliminate tedious tasks and boost personal productivity.
Social effects and community
They also worry about social consequences. Agentic tools that simulate collaboration might reduce real human interaction. Nathan and Robby hope for balance: digital tools should augment work and creativity without replacing in‑person community. They miss WordCamps and the bonds formed by travel and meetups, noting a decline in events since the pandemic and expressing hope for a revival.
Closing perspective
Beaver Builder aims to integrate AI thoughtfully—skipping hype, focusing on practical features that remain transparent, and enabling users to learn from and tweak generated output. Page builders will likely continue to serve as important editors and maintenance layers, while WordPress’s extensibility remains a major advantage for complex or long‑term projects. Both hosts finish with cautious optimism: excited about new capabilities, aware of cultural tradeoffs, and hopeful that craftsmanship and face‑to‑face community will endure alongside evolving tools.