# Artificial Intelligence Journal

> A personal reader for AI news, papers, model releases, markets, and discussion — sourced from primary feeds.

This index is generated from `lib/site.ts` and `lib/explainers.ts`. The canonical site is at https://ai-news-brief-neon.vercel.app.

## Sections

- [The Brief](https://ai-news-brief-neon.vercel.app/): Home — markets, model leaderboard, news, YouTube, companies, papers, releases, discovery.
- [News](https://ai-news-brief-neon.vercel.app/news): Wire-style AI and tech news from primary publisher feeds.
- [YouTube](https://ai-news-brief-neon.vercel.app/youtube): Curated AI talks, demos, and long-form interviews.
- [Companies](https://ai-news-brief-neon.vercel.app/companies): Announcements direct from AI labs, infra vendors, and big-tech engineering teams.
- [Discovery](https://ai-news-brief-neon.vercel.app/discovery): Top stories from Hacker News — community signal.
- [Papers](https://ai-news-brief-neon.vercel.app/papers): Newest AI papers from arXiv and HuggingFace Daily Papers.
- [Releases](https://ai-news-brief-neon.vercel.app/releases): Newly available AI models on OpenRouter, with context, modality, and pricing.
- [Explainers](https://ai-news-brief-neon.vercel.app/explainers): Long-form mechanism-level pieces on how AI systems actually work.

## Explainers

- [What is a Large Language Model?](https://ai-news-brief-neon.vercel.app/explainers/what-is-an-llm): The basic idea behind models like GPT and Claude.
- [Tokens & Embeddings](https://ai-news-brief-neon.vercel.app/explainers/tokens-and-embeddings): How text becomes numbers a model can work with.
- [The Transformer Architecture](https://ai-news-brief-neon.vercel.app/explainers/the-transformer): The design that made modern AI possible.
- [Pretraining vs Fine-tuning](https://ai-news-brief-neon.vercel.app/explainers/pretraining-vs-finetuning): The two-stage recipe behind most chat models.
- [RLHF Explained](https://ai-news-brief-neon.vercel.app/explainers/rlhf): How human preferences steer a model's answers.
- [Prompting Basics](https://ai-news-brief-neon.vercel.app/explainers/prompting-basics): Getting better answers without changing the model.
- [Retrieval-Augmented Generation](https://ai-news-brief-neon.vercel.app/explainers/rag): Giving a model fresh, private knowledge at answer time.

## Data feeds

- [Full RSS feed](https://ai-news-brief-neon.vercel.app/feed.xml): Aggregated journal feed across all sections.
- [Sitemap](https://ai-news-brief-neon.vercel.app/sitemap.xml): Machine-readable list of every routable page.
- [Full corpus](https://ai-news-brief-neon.vercel.app/llms-full.txt): Expanded Markdown with the full explainer bodies inlined.

## Upstream sources

- [Artificial Analysis](https://artificialanalysis.ai/): Intelligence index, coding/agentic benchmarks.
- [OpenRouter](https://openrouter.ai/): Model leaderboard, pricing, and new release feed.
- [Hugging Face](https://huggingface.co/): Daily Papers feed.
- [arXiv](https://arxiv.org/): Newest cs.AI/cs.LG papers.
- [Hacker News](https://news.ycombinator.com/): Discovery section.
- [Polymarket](https://polymarket.com/): Prediction markets.
- [Finnhub](https://finnhub.io/): Equity quotes.
- [YouTube Data API](https://developers.google.com/youtube/v3): Talks and interviews.