Tag: llm
Everything tagged “llm” — my own articles first, then the AI Radar entries that carry the same tag.
AI Radar
- · generativeai.pub
Memory Isn’t a Bigger Context Window
Learn why larger context windows fail as memory solutions in AI systems. Discover how to design governance layers for validated knowledge, expiration rules, and trust models instead of simple retrieval augmentation.
- · blog.lepine.pro
When code complexity lives in the prompt
Discover why traditional code metrics fail for LLM apps and learn new ways to measure prompt complexity based on maintenance effort correlations from recent research.
- · generativeai.pub
This Repo Cut My Agent’s Token Bill by 88% and the Answer Didn’t Change
Discover how the headroom tool reduces LLM token costs by compressing tool outputs and logs before model ingestion using reversible compression techniques.
- · technologyreview.com
Three reasons why DeepSeek’s new model matters
Explore DeepSeek V4's impact on AI development, covering open-source architecture, 1M token context efficiency, and shift toward domestic chip infrastructure.
- · generativeai.pub
Context Engineering Is the New Moat
Discover why context engineering is the key differentiator for AI products, surpassing model selection. Learn about its disciplines and long-term competitive advantages.
- · generativeai.pub
The New LLM risk : Skills
Explore a new LLM attack vector leveraging malicious skills in platforms like Claude. Learn how attackers can exploit user-installed skills to compromise AI systems.
- · generativeai.pub
This Simple CLAUDE.MD File Went Viral with 130K GitHub Stars
Discover how a simple CLAUDE.md file gained 130K GitHub stars by providing guidelines for AI coding agents to improve code quality and reduce common mistakes.
- · blog.cloudflare.com
Project Glasswing: what Mythos showed us
Cloudflare's Project Glasswing uses Anthropic's Mythos Preview to find vulnerabilities. Learn about exploit chain construction, model limitations, and harness design.
- · generativeai.pub
RAG Is Fundamentally Broken. Here’s Why.
Explore the limitations of Retrieval-Augmented Generation (RAG) systems, why chunk tuning is insufficient, and how Apple's CLaRa offers a potential solution through differentiable retrieval.
- · generativeai.pub
Everything You Need to Know About RAG
Explore Retrieval-Augmented Generation (RAG): how it works, its architecture (ingestion & retrieval pipelines), chunking/retrieval strategies, and limitations.