Blog
Should You Actually Use Redis Iris? An Honest Builder's Verdict
Redis Iris solves real production agent problems. It also requires real operational commitment. Here is the 5-question checklist I use to decide.
Redis Iris vs. Pinecone Nexus vs. Naive RAG: When to Use What
The only question that matters when choosing a retrieval architecture: how fast does your data change? Here is the full comparison with three concrete scenarios.
Why Your RAG Pipeline Fails in Production
Most production RAG failures are not model problems. They are runtime data problems. Here is what actually breaks and why better retrieval does not fix it.
How Redis Iris Actually Works: A Builder's Breakdown
Redis Iris has four moving parts. Here is how RDI, the Context Retriever, Agent Memory, and LangCache actually work, with honest tradeoffs on each.
We Cut 63% of Our LLM Costs by Sending Less Context
We moved from 128K to 32K context windows. Costs dropped 63%. Accuracy went up. Here is how metadata filtering, re-ranking, and dynamic budgets do it.
The LLM Wiki Pattern: How to Build a Personal Knowledge Base That Actually Compounds
Andrej Karpathy's LLM Wiki pattern replaces RAG with a persistent, LLM-maintained wiki that gets smarter with every source you add. Here is how it works and why it matters for AI engineers.