MCP Tasks, introduced in the 2026-07-28 Model Context Protocol specification, allow servers to respond to tool calls with a durable task handle instead of a blocking result, enabling context offloadin…
A comprehensive guide published in 2026 details 25 retrieval-augmented generation (RAG) strategies, organized into five pipeline layers, to help engineers and researchers move beyond naive RAG impleme…
A study by Chroma found that 65% of enterprise AI failures in 2025 are caused by context drift or memory loss during multi-step reasoning, not model capability issues. Researchers propose a three-laye…
Chroma tested 18 frontier AI models including GPT-4.1, Claude Opus 4, and Gemini 2.5 Pro, finding that every model's performance degrades as context length grows, contradicting vendor claims of 1M-2M …
A new analysis warns that large language model context windows degrade significantly beyond 100k tokens, making advertised sizes of 200k to 2M tokens misleading for practical use. Studies like RULER a…
A developer explains that RAG (Retrieval-Augmented Generation) reduces LLM hallucinations by fetching relevant document chunks at query time and instructing the model to answer using only that context…
A developer built a RAG (Retrieval-Augmented Generation) system to let their team chat with internal documentation, replacing a wiki search that failed to answer common questions. After experimenting …
A developer built a persistent memory system for multi-turn RAG chatbots using ChromaDB, solving the problem of conversation history loss after service restarts. The solution stores conversation turns…
Patrick Jiang (@patpcj) released Harness-1, a 20 billion parameter search agent that externalizes search state from the model into a structured harness, as detailed in a 13-post thread on X and an acc…
A developer's guide to CLAUDE.md files recommends keeping them under 200 lines with only universally applicable rules, as bloated files degrade model accuracy from 95% to 60% based on a 2025 Chroma st…
A developer built RepoChat, an AI tool that uses retrieval-augmented generation (RAG) to answer questions about GitHub repositories. The tool indexes codebases by filtering relevant files, chunking co…
A developer built a fully local Retrieval-Augmented Generation (RAG) system using Ollama and TypeScript, requiring no API keys or third-party calls. The 200-line command-line tool indexes `.md` and `.…
To build a production-ready Retrieval-Augmented Generation (RAG) pipeline in Python using BM25 retrieval via Meilisearch instead of a vector database, arguing that BM25 achieves 85–95% of the recall o…
"RAG Readiness," a tool designed to eliminate the vague "it depends" answers that plague Retrieval-Augmented Generation (RAG) architecture decisions. Instead of providing comparison tables, the tool u…
To implement retrieval-augmented generation (RAG) using the pgvector extension within an existing PostgreSQL database, combined with Amazon Bedrock's Titan embedding model and Claude for answer genera…
Production RAG systems often fail after moving beyond demo stage due to underbuilt indexing, retrieval, and observability layers. The indexing pipeline ingests documents into chunks and vector embeddi…
Public benchmarks like MTEB are unreliable for evaluating embedding models in real-world retrieval systems because their generic, artificially clean data does not reflect actual user behavior or domai…