I built an AI chef for my family
The hardest part at home isn't cooking, it's deciding what to eat. I built a small module that recommends tonight's dish from a photo of the fridge. A development note for something not yet shipped.
#LLM
Total 4 posts
Filtering noise from RAG chunks before they poison retrieval
After splitting a document, not every chunk is worth keeping. Ads, nav bars, disclaimers, and duplicates are noise that pollutes retrieval. This walks rule-based, structural, semantic, and LLM filtering strategies, with a Java pipeline you can adapt.
Xuelao dialect is disappearing, so I built this app
Xuelao is a Hoklo dialect around Shanwei, with no standard writing, no teaching resources, and young people no longer speaking it. I built a translation and learning app — harder than I expected.
RAG PDF chunking: full text into the base or extract fields? A counterintuitive conclusion
The first decision in RAG isn't picking a vector DB, it's what you store after chunking the PDF. Full-text ingestion and structured extraction are two completely different paths; data from a medical scenario makes this very clear.