UHUpadrasta Harshainharshaupadrasta.hashnode.devยท2d ago ยท 14 min readToken-Efficient Adaptive RAG: Building a RAG System That Knows What NOT to RetrieveHow I built an adaptive Retrieval-Augmented Generation platform that reduces embedding and context tokens without blindly sacrificing retrieval quality Introduction Retrieval-Augmented Generation (RA00
UHUpadrasta Harshainharshaupadrasta.hashnode.devยทMar 12 ยท 7 min readDay 5 โ Chunking: The Secret to Better RAG Performance ๐งฉ๐ค๐ Advanced AI Engineering Series Artificial Intelligence systems are becoming incredibly powerful at answering questions, summarizing documents, and assisting with complex tasks. However, the effecti00
UHUpadrasta Harshainharshaupadrasta.hashnode.devยทMar 11 ยท 6 min readDay 4 โ How Retrieval Works Inside a RAG System๐ Advanced AI Engineering Series In the previous articles of this series, we explored two powerful building blocks behind modern AI systems: ๐ง Embeddings โ which convert text into numerical vectors 00
UHUpadrasta Harshainharshaupadrasta.hashnode.devยทMar 10 ยท 5 min readDay 3 โ Why Vector Databases Are Critical for AI๐ Advanced AI Engineering Series In the previous article, we explored embeddings, the hidden backbone of RAG systems. Embeddings convert text into numerical vectors that represent meaning, allowing A00
UHUpadrasta Harshainharshaupadrasta.hashnode.devยทMar 9 ยท 4 min readDay 2 โ The Hidden Backbone of RAG: EmbeddingsIn the previous article, we explored why modern AI systems need RAG (Retrieval Augmented Generation). RAG allows AI systems to retrieve relevant information from trusted knowledge sources before gener00