AI doesn't have to replace Drupal
I see AI primarily as an additional layer on top of a strong CMS. Drupal provides structure, content, permissions, and integrations. AI can help generate, understand, transform, and unlock that content. It is precisely this combination that creates interesting possibilities.
Content
AI can help generate, rewrite, summarize, translate, and structure content. I mainly look at applications that fit into existing editorial workflows and where AI genuinely saves time or improves quality.
Development
I use AI as a programming partner for research, prototypes, code, tests, debugging, and refactoring. This allows me to try out ideas faster and spend more time on architecture, technical decisions, and solving the interesting problems.
Interaction
AI enables new ways of working with Drupal content. Think conversational interfaces, AI assistants, smart search functions, and ways to help visitors navigate large amounts of content more naturally.
Automation
Where people repeatedly perform the same actions, I look at whether AI can take over part of that work. Not to automate processes for the sake of automation, but to explore where human time can be put to better use.
Vibe coding, talking to your codebase
One of the most interesting developments I see is vibe coding. You no longer always have to start with a fully worked-out technical plan. You describe what you want to build, explore together with AI what is needed, and then build step by step.
For example:
"I want an interactive component on this Drupal website where visitors can ask a question and receive relevant content as a result."
From there, I can quickly create an initial prototype with AI. Not to make the developer redundant, but precisely to enable faster thinking, building, and testing.
AI is my sparring partner, programming assistant, and accelerator for the tedious work. The architecture, technical decisions, review, and responsibility remain with me.
From idea to prototype
As a freelancer, I can start relatively small. No months-long process before anything becomes visible, but rather taking an idea together and seeing how far we can take it.
01 — Idea
We determine what you actually want to achieve and what problem you are trying to solve.
02 — Experiment
I investigate what Drupal and AI can contribute and where the interesting possibilities lie.
03 — Prototype
We quickly build something that lets you experience the idea instead of just talking about it.
04 — Testing
We look at what works, what doesn't, and where real value is created.
05 — Build
If the idea works, we turn it into a solid Drupal solution that fits within your existing platform and technical environment. I carefully review all AI-generated code for coding standards, scalability, security, and all Drupal best practices.
06 — Iterate
Continuous development with AI, including integrating and reviewing the relevant 'vibe code'.
What should you pay extra attention to in Drupal?
Vibe coding is especially powerful when the developer continues to carefully follow Drupal best practices.
For example, pay attention to:
- Security — check input, permissions, and output.
- Drupal coding standards — do not automatically assume generated code is correct.
- Upgradability — verify that the APIs being used actually match the Drupal version.
- Dependency injection — avoid unnecessary use of global services.
- Configuration management — make sure settings are exportable.
- Cacheability — blocks and entity queries must be cached correctly.
- Access checking — for all 'Routes' and 'Entity Queries'.
- Render arrays — avoid unnecessary HTML through
#markup. - Twig escaping — let Drupal handle escaping of output.
- .. and many other Drupal best practices
Additional Drupal development
I build new functionality, custom modules, theming, and integrations, and help further develop existing Drupal platforms.
AI for content
I explore ways to make content workflows smarter and more efficient, without taking control away from editorial teams and content teams.
Technical experiments
Do you have an idea but don't yet know whether it is technically possible or will actually provide value? We can build and test it first before turning it into a major project.
No AI for the sake of AI
I'm not interested in putting a chatbot on every page. I'm interested in the question: can we actually make this better with AI?
Sometimes the answer is yes. Sometimes a regular Drupal solution is better. And sometimes the most interesting result is a prototype that shows why you shouldn't build something.
I find that experimentation interesting.
Drupal × AI × me
I combine my Drupal experience with new AI tools and an experimental way of working. So you don't need to know exactly what is technically possible.
Bring an idea, a problem, or even just a "Wouldn't it be cool if…".
Then let's see what we can build together.