5 min read
How RAG Actually Works (It's Mostly Retrieval)
A plain-English guide to how RAG works. Retrieval-augmented generation is really a prompt trick, and the retriever is where your system succeeds or fails.
- RAG
- LLMs
- AI Engineering
Blog
Writing & notes.
Thoughts on production AI/RAG systems, LLMs, agentic AI, and full-stack engineering.
5 min read
Hybrid Search: Vector Search Alone Isn't Enough
Semantic search quietly misses exact terms. Here's why hybrid search, combining BM25 keyword search with vector search plus RRF, gives RAG far better retrieval.
- RAG
- Vector Search
- AI Engineering
6 min read
Hydration Explained: SSR, CSR, and Why It Breaks
A plain-English guide to hydration in React and Next.js: what CSR and SSR really do, why hydration errors happen, and how to fix the server-client mismatch.
- Next.js
- React
- Web Development
2 min read
Building Production RAG Systems That Actually Hold Up
What separates a RAG demo from a production system: retrieval quality, grounding with citations, and real evaluation.
- RAG
- LLMs
- Architecture