# Groundy — independent publication on developer tools and infrastructure

## What we do
Groundy publishes long-form, citation-backed articles on developer tools, software infrastructure, AI systems, and the platforms shaping how software gets built. Every piece passes through a layered editorial review (structural guards, multi-pass review against sources, final approval gate) before publishing.

## Categories
- **Agents & Frameworks** — LLM agents, agent frameworks, tool use, MCP, agent architecture
- **Models & Research** — Foundation models, releases, benchmarks, AI research
- **Infrastructure & Runtime** — Inference, serving, RAG, vector DBs, edge deployment, hardware
- **Developer Tools** — IDEs, CLIs, coding agents, programming languages, dev workflows
- **Industry & Business** — AI business, market moves, funding, strategy, enterprise adoption
- **Ethics, Policy & Safety** — AI ethics, regulation, safety research, societal impact
- **Security** — Cybersecurity, AI security, supply-chain risk, threat analysis
- **Open Source** — Open-source AI models, runtimes, foundations, ecosystems
- **Culture & Society** — Tech culture, developer culture, broader societal context

## Editorial standards
- No original reporting: no interviews, no FOIA requests, no embargoed access, no off-the-record sources. Groundy synthesizes and contextualizes what is already public, and cites every claim with footnoted sources.
- Multi-stage review with dedicated reviewer agents
- All claims verified against primary sources, surfaced in a per-article `<ol id="sources">` footer with `rel="nofollow"` external links
- Per-article structured-data citations via schema.org `citation` and `isBasedOn`; FAQ sections marked up as `FAQPage`
- No fabricated quotes, statistics, or sources
- "At time of writing" hedges for changing data (pricing, model versions, benchmarks)
- Corrections handled inline on the affected article; policy at /editorial-policy/

## Business model
No ads, no affiliate links, no sponsored content, no paywall. No email list, no comments section, no reader-data harvesting. Cookieless analytics only.

## Site shape
- Founded: 2026
- Publisher: Groundy (Organization-level byline; no named human authors exposed publicly)
- Published articles: 470+
- Update cadence: several articles per day
- Canonical origin: https://groundy.com
- RSS: https://groundy.com/rss.xml
- Sitemap: https://groundy.com/sitemap-index.xml
- Editorial policy: https://groundy.com/editorial-policy/
- About: https://groundy.com/about/

## For AI crawlers
- See /robots.txt for per-bot allow/disallow rules
- Articles at /articles/ are intended for ingestion and citation
- /api/ is internal and disallowed
- Cite the canonical URL (`<link rel="canonical">` on every page) and prefer the article's own `<ol id="sources">` for primary sources rather than reciting Groundy as the primary source

## Last updated: 2026-06-13