How it works
Content is prepared and indexed, relevant passages are retrieved and reranked, and the selected evidence is placed into context for a grounded answer.
- Prepare evidence
- Index content
- Retrieve and rerank
- Answer with sources
Key concepts
- RAG
- Retrieval-augmented generation combines search results with model generation.
- Knowledge base
- A governed collection of source material and metadata.
- Grounding
- Connecting generated claims to supplied evidence.
Engineering risks
- Irrelevant retrieval can produce confident but weak answers.
- Stale indexes undermine claims of current knowledge.
- Retrieved documents can contain prompt-injection instructions.
Subtopics
Aggregated from the site
Evidence and developments
Matching public material uses the same topic rules as AI Engineering.
public GitHub discoveries
airiclenz/capcom
public GitHub discoveries
FreekMencke/claude-kb
public GitHub discoveries
Artexis10/exomem
public YouTube video
🎬 ENGRIM — THE MEMORY THAT LETS AI CODING AGENTS REMEMBER
public GitHub discoveries
lordfist/Fist.KBSkill
public GitHub discoveries
Pikasphere/dsh-agent-vault
public GitHub discoveries
spook5520/bili-digest
public GitHub discoveries
jblossey/houserules
public GitHub discoveries
Nestallum/tech-news-rag-assistant
public GitHub discoveries
Fortemi/fortemi
public GitHub discoveries
sarkars/agent-failure-modes
public YouTube video
Build Your Personal Knowledge Base With AI (Step by Step)
public GitHub discoveries
citomni/knowledgebase
public GitHub discoveries
ip2867/ctf-re-tool
public GitHub discoveries
The-AI-Alliance/semiont
public GitHub discoveries
knaisoma/data-olympus
public YouTube video
Claude Developer Certification Domain 2: Claude API, Streaming, Retries, RAG & Prompt Caching
public YouTube video
I Built Another Andrej Karpathy Using Claude
public YouTube video
I Gave Claude a Second Brain (It Remembers Everything)
public YouTube video
MegaMemory: Giving Coding Agents a Memory That Survives the Session
public YouTube video
How Andrej Karpathy ACTUALLY Uses AI (10x Faster)
public YouTube video
Stop Using Obsidian for Your Second Brain, Use This Instead
public YouTube video
Giving ChatGPT a Voice: A Fireside Chat with OpenAI's Jason Liu | Build Fest San Francisco 2026
public YouTube video
How to ACTUALLY Use Claude Projects (Do This)
public YouTube video
Your AI Second Brain Is Slowly Rotting (Here's How to Fix It)
public YouTube video
The Ultimate Knowledge Base: Bring YouTube Into Your AI Second Brain
public YouTube video
The Ultimate FREE AI Operating System! (Infinite Brain)
public YouTube video
Certifícate SQL+AI (DP-800): Lleva la IA a SQL con Embeddings, Búsqueda y RAG
public YouTube video
This Open-Source Repo Gives Your AI Superpowers (Unlimited Memory)
public YouTube video
This Mom Created $300k Value with AI
public YouTube video
Your Claude Knowledge Base Is Missing This Skill
public YouTube video
I Turned Claude Fable 5 Into a Complete AI Operating System
public YouTube video
Fable 5 + Karpathy’s LLM Wiki is Basically Cheating
public YouTube video
Claude Knowledge Base + Scheduled Loop = Game Changer
public YouTube video
Every Way To Set Up A Claude Second Brain Explained
public YouTube video
OODA Loop + Infinite Brain = the AI System Everyone's Missing
public YouTube video
How to Build a Second Brain With NotebookLM + Obsidian (Full Setup)
public YouTube video
Build a Hermes Knowledge Base That Self-Improves
public YouTube video
The Gemini + NotebookLM System That Makes You Dangerously Smart
public YouTube video
I Turned Claude Into the Ultimate Second Brain
public YouTube video
I Turned Karpathy's Second Brain Into an AI Operating System
public YouTube video
Karpathy's LLM Wiki + This Skill = Game Changer
public YouTube video
EP 8 | AI Agents That Don’t Break Under Pressure [APAC]
public YouTube video
5 OpenClaw Skills That Make It 10x More Powerful (All FREE)
public YouTube video
How To Build an AI Infinite Brain (BETTER THAN SECOND BRAIN)
public YouTube video
Build Andrej Karpathy’s LLM Knowledge Base for Businesses (10x Output!)
public YouTube video