29-01-2025
AI is rapidly transforming industries, and Magento 2 development is no exception. With large language models (LLMs) improving at an astonishing rate, automating parts of the development workflow is no longer a futuristic dream—it’s happening now.
To explore the capabilities of AI in Magento 2 development, we trained an LLM using approximately 1GB of Magento 2 and Hyvä documentation. This provided the model with a deep understanding of the framework, its structure, and best practices.
Once the model was trained, we built an AI agent around it and integrated it into our GitLab workflow. The goal? To automate the process from issue creation to deployment—minimizing the need for human intervention while maintaining quality control.
This is the flow for how the AI agent works. We will describe the steps further below.
1. Client Opens an Issue
The process begins when a client submits a new issue. This could be a request for a bug fix, a small feature, or another change within the Magento 2 environment.
2. AI Agent Takes Over
The AI agent analyzes the issue, generates code changes, and opens a merge request. It then commits the changes to a dedicated branch and triggers a GitLab pipeline.
3. Automated Review Environment
The pipeline spins up a review environment, allowing the client to verify the AI-generated changes before a developer gets involved.
4. Developer Code Review
If the client is satisfied with the update, a developer steps in to review the merge request, ensuring the changes meet our coding standards and business requirements before merging.
5. Deployment of Changes
After approving and merging the merge request, the code is deployed to the live environment.
Right now, the AI agent handles simple development tasks effectively. It can generate straightforward Magento 2 and Hyvä code modifications, such as:
However, I’m continuously testing ways to improve its capabilities. Currently, it runs on Llama 3.3 with the Magento 2 and Hyvä dataset, but I’m planning to test DeepSeek R1 alongside it to compare results. Over time, we expect it to take on more complex tasks, potentially handling larger feature implementations and deeper integrations.
While AI won't replace developers, it is already proving to be a valuable assistant. Automating routine tasks allows developers to focus on more complex problems, architecture decisions, and high-level optimizations. As AI models continue to evolve, we could see them taking on increasingly sophisticated responsibilities.
The potential is huge, and this AI-driven workflow is just the beginning. By integrating AI into Magento development, we’re not only making our workflows more efficient but also redefining how development teams interact with clients and their requests.
Stay tuned for further updates as we refine this AI agent and push the boundaries of Magento 2 automation!
