Programming with an AI Agent
Artificial intelligence (AI) has fundamentally changed the way we program. AI agents can generate code, optimize it, and even assist with debugging. However, there are some limitations that programmers must keep in mind when working with AI.
Issues with sequence and duplication
AI agents struggle with the correct order of code. For example, they may place initializations at the end of a file, which causes runtime errors. Additionally, AI may unhesitatingly define multiple versions of the same class or function within a project, leading to conflicts and confusion.
A code platform with memory and project structure helps
One solution to this is using AI code platforms that can manage memory and project structures. This helps maintain consistency in complex projects. Unfortunately, these features are not always applied consistently. As a result, the AI may lose the coherence of a project and introduce unwanted duplications or incorrect dependencies during programming.
Most AI coding platforms work with so-called tools that the large language model can invoke. These tools are based on an open standard protocol (MCP). It is therefore possible to link an AI coding agent to an IDE such as Visual Studio Code. Optionally, you can set up a local LLM with Llama or Ollama and choose an MCP server to integrate with. NetCare has created a MCP server designed to help with debugging and managing the underlying (Linux) system. Useful if you want to deploy code live immediately.
Models can be found on Hugging Face.
IDE extensions are indispensable
To better manage AI-generated code, developers can use IDE extensions that monitor code correctness. Tools such as linters, type checkers, and advanced code analysis tools help detect and correct errors early. They form an essential complement to AI-generated code to ensure quality and stability.
The cause of recurring errors: context and role in APIs
One of the main reasons AI agents keep repeating mistakes lies in the way AI interprets APIs. AI models need context and a clear role definition to generate effective code. This means prompts must be complete: they must not only contain the functional requirements, but also explicitly state the expected outcome and constraints. To facilitate this, you can save prompts in a standard format (MDC) and send them along with the AI by default. This is especially useful for generic programming rules you apply, as well as the functional and technical requirements and the structure of your project.
Tools like FAISS and LangChain help
Products such as FAISS and LangChain offer solutions to help AI handle context better. For example, FAISS assists in efficiently searching and retrieving relevant code snippets, while LangChain helps structure AI-generated code and maintain context within a larger project. But here too, you can optionally set it up locally yourself using RAG databases.
Conclusion: useful, but not yet autonomous
AI is a powerful tool for programmers and can help accelerate development processes. However, it is not yet truly capable of independently designing and building a more complex codebase without human oversight. Programmers should view AI as an assistant that can automate tasks and generate ideas, but still requires guidance and correction to achieve a good result.
Contact contact us to help set up the development environment, assist teams in getting the most out of it, and focus more on requirements engineering and design rather than debugging and writing code.
