#RAG

8 articles tagged with RAG

Context Window Wars: Do You Actually Need a Massive Context Model?

2M-10M token context windows vs retrieval accuracy. Why 'Needle in a Haystack' decay, prefill latency penalties, and financial token costs make RAG + prompt caching the smarter 2026 choice.

8/1/202623 min read

Building a Production RAG API with FastAPI and pgvector

A complete build of a retrieval-augmented generation endpoint on top of Postgres and pgvector — chunking, indexing, hybrid search, and the reranking step most tutorials skip.

7/20/202613 min read

From RAG to Agentic Retrieval: How Retrieval Patterns Evolved in 2026

A single embed-and-retrieve step was never going to be the final form of grounding an LLM in real data. Here's how retrieval architectures actually changed as teams ran into its limits.

7/19/202612 min read

Hierarchical Document Retrieval: Designing Cache-Optimized RAG Datasets

Optimize RAG queries. Learn how to structure document trees to maximize context cache hits and reduce token consumption.

7/10/202612 min read

RAG Optimization: Implementing Local Embedding Cache Layers

Speed up Retrieval-Augmented Generation (RAG). Learn how to implement an IndexedDB-based embedding cache to eliminate redundant network API calls.

7/10/202612 min read

Building a Local RAG Pipeline inside the Browser with SQLite-VSS and WebGPU

Learn how to build a fully client-side Retrieval-Augmented Generation (RAG) pipeline. Query documents locally using SQLite-VSS vector search and WebGPU-accelerated LLMs.

6/4/202612 min read

Edge-Native Search: Implementing Local RAG in the Browser

The future of search is personal, private, and fast. Learn how to build a Retrieval-Augmented Generation (RAG) system that runs entirely on the client, using WebGPU and Vector DBs.

4/15/202616 min read

The Death of the Search Engine: How LLMs Rewrote the Information Web in 2026

Searching for keywords is a relic of the past. Discover how LLMs and agentic RAG have fundamentally changed how we find and consume information in 2026.

4/6/202612 min read