Can #AI coding assistants handle legacy code without breaking everything? @lizeraes compares the tools that refactor safely vs. those that hallucinate fixes.
Read before you trust them: https://javapro.io/2025/07/02/ai-tooling-for-software-development-the-useless-the-useful-and-the-future/
#Frameworks #DevOps #LLM #DevoxxGenie #LangChain4j @langchain4j
#GitHubCopilot, Cursor, #DevoxxGenie: same promise, different results. @lizeraes tested leading #AI coding tools on real workflows. The gaps between them are bigger than you think.
Read the detailed comparison: https://javapro.io/2025/07/02/ai-tooling-for-software-development-the-useless-the-useful-and-the-future/
#Frameworks #LLM #LangChain4j @langchain4j
Now experimenting LLMs as a code assistant in my IDE, at last! π§βπ»+π€/π§
Using #Ollama and #DevoxxGenie #IntelliJ_IDEA plugin on my 10-year old macOS laptop! π¦/π
https://github.com/devoxx/DevoxxGenieIDEAPlugin
Seems easier that copy/pasting code blocks in Chat GPT. π
DevoxxGenie is a plugin for IntelliJ IDEA that uses local LLM's (Ollama, LMStudio, GPT4All and Llama.cpp) and Cloud based LLMs to help review, test, explain your project code. - devoxx/DevoxxGe...
JAVAPRO
Stephan Janssen
Nicolas Kosinski