Coding & Development
RAG.md · 26 documents
plan
* **Embedding Model:** Switched from a generic model to the specific `EmbeddingGemma` model mentioned in the documentation.
📝 Development Notes
I used Claude/GPT assistance for the following components, which I then extensively modified and enhanced:
🚀 Retrieve — Implementation Plan
> **Full-stack RAG (Retrieval-Augmented Generation) Application**
Socrates Development Plan
- Create two top-level folders: `backend/` and `frontend/`
Building a RAG Pipeline from Scratch
**Retrieval-Augmented Generation (RAG)** is the technique behind most production AI assistants today. Instead of asking an LLM to recall facts from training, you give it relevant context at query time. The result is more accurate, grounded, and up-to-date answers.
Multilingual Agentic RAG (venv, Qdrant, NiceGUI, Gemini-001)
- venv only; Windows-first; no WSL
The Simpsons Episode Database - Technical Plan
A local application that scans TV episodes of The Simpsons, transcribes them, creates searchable embeddings, and provides a web interface for semantic search and video clip creation.
DinAI - RAG AI Chatbot Project Plan
DinAI is a Retrieval-Augmented Generation (RAG) AI chatbot that provides intelligent responses based on scraped web content stored in a vector database.
Partie 2 : Pipeline RAG
Mettre en pratique la conteneurisation et l'orchestration d'une application de Génération Augmentée par Récupération (RAG) à l'aide de Docker. Ce TP vous permettra de comprendre comment déployer une application d'intelligence artificielle moderne en utilisant les meilleures pratiques DevOps.
RAG (Retrieval Augmented Generation) Implementation - Technical Outline
- Combines custom data with LLMs
Implementation Plan: codeindex
We're building a local semantic index for codebases that augments Claude Code's Glob/Grep/Read tools with embedding-based search. The design is informed by OpenViking (AST skeletons, bottom-up directory summaries, score propagation), Augment Context Services (single retrieval tool for agents), sigma-ralph-grindset (`claude --print` as LLM backend), and db-harness (Bun-native dual-database skill pattern).
Loop #1 – RAG Pipeline
Building a YouTube video RAG pipeline for an AI coach agent. The pipeline pulls transcripts from a specific YouTube channel, chunks them, embeds them, and stores them in a vector database for agent retrieval.
packrat (development version)
- Update vendored `renv` to support recognition of Posit Package Manager
On Mac/Linux
`YOUR_OPENAI_API_KEY` or `YOUR_VECTOR_DB_CONFIGURATION` fill in the details specific to your setup.\*\*
RAG-Lighter Wiki
> 轻量级模块化 RAG (检索增强生成) 框架完整文档
How to set up QMD for Claude
Here’s the full setup, step by step.
Development Notes - AI Agent Server
- **Initial project structure and boilerplate**: Used AI to scaffold the basic Express.js server setup
Development Diary
Building a Document Similarity Server with CLI interface that:
Markdown VDB — Technology Decisions
Rust toolchain mapped to each functional area of the project. Every choice justifies **why this crate** for **this requirement**.
COSMIC Command Line Interface
Complete reference for the `cosmic` CLI tool.
Technical Documentation - NanoScope
- 🔗 **GitHub Repository**: [github.com/sagred/nano](https://github.com/sagred/nano)
extend_retrieval
Extending Retrieval in Terrier
softrag
**softrag** is a minimalist local-first Retrieval-Augmented Generation (RAG) library that uses SQLite with [sqlite-vec](https://github.com/asg017/sqlite-vec) for efficient storage of documents, embeddings, and cache in a single `.db` file.
Quickly build Generative AI applications with Amazon Bedrock
This repository contains code samples for building diverse AI applications using Amazon Bedrock's foundation models. Learn how to accelerate projects in image and text generation and beyond.