Posts
Beyond Traditional RAG: Fixing a Core Limitation with an Agent That Asks Clarifying Questions
**Repository:** 👉 [premsgdev/rag-engine](https://github.com/premsgdev/rag-engine) Retrieval-Augmented Generation (RAG) is widely used to reduce hallucinations in Large Language Models (LLMs). The idea is simple: retrieve relevant documents and force the model to answer only from them. But while RAG helps with hallucination, it still fails in one critical area: **Ambiguous user queries.** In this article, I’ll explain: * Why traditional RAG breaks on ambiguity * Why vector similarity alone is not …
Building a Hybrid RAG Engine with Local + Cloud Embeddings
Retrieval-Augmented Generation (RAG) is one of those patterns that becomes more interesting the deeper you go. Once you begin working with real workloads, two challenges show up quickly: - Embedding API calls become **expensive** - Cloud latency can be **unpredictable** I wanted to experiment with a setup that balances cloud power with local flexibility—a system that feels good to develop with and doesn’t punish you for iterating. This led to a small project I’ve been improving over time: 👉 …
Modern WordPress with Docker + Bedrock
WordPress powers a huge portion of the internet, but its **default development workflow** hasn’t kept up with modern engineering standards. No dependency management, mixed configuration files, plugins committed to Git, and environments that behave differently on each machine. To solve these issues for my own workflow, I built **[wp-bedrock-container](https://github.com/premsgdev/wp-bedrock-container)** — a modern WordPress setup powered by **Bedrock**, **Docker**, **Composer**, **Nginx**, and …