citability.dev
AI Visibility Auditing
AI visibility auditing platform. 10 verified infrastructure checks measuring whether AI systems can find, recommend, and cite your website.
chudi.dev is a live public experiment in AI-visible web architecture. It combines technical writing, machine-readable discovery, and agent-facing interfaces so the same body of work can be browsed, retrieved, cited, and queried.
AI-visible web architecture is the practice of structuring a website so it serves three audiences simultaneously: human readers through editorial design, large language models through machine-readable content surfaces (llms.txt, structured data, FAQ schema), and AI agents through callable tool interfaces (WebMCP, ai.txt). This site is the reference implementation.
Visibility stack
chudi.dev is designed so readers, search engines, and AI systems can all navigate the same authority graph through interfaces tuned to how they retrieve information.
AAO
WebMCP tools, ai.txt, and callable content surfaces make chudi.dev usable by agents, not just readable by them.
AEO
Definition blocks, FAQ structure, source sections, and reading paths turn posts into extractable answers instead of generic essays.
GEO
llms.txt, structured data, sitemaps, and entity clarity make the site legible as a coherent knowledge node instead of a generic content archive.
Three editorial journeys
Each journey is a topic hub, a recommended reading order, and a newsletter segment. The goal is faster orientation and stronger follow-through.
This track is the engineering playbook for the second game: entity coherence, schema specificity, originality signals, and the measurement loops that show whether ChatGPT, Perplexity, and Google AI Overviews are actually citing you.
Best for
content engineers and sub-DR-20 operators engineering for AI citations
Reading depth
16 foundational, tactical, and case-study notes
Start here
Answer Engine Optimization: 6 Factors That Decide If AI Cites You
Answer engine optimization (AEO) determines which sites AI search engines cite. The 6 factors driving citations in Perplexity, ChatGPT, and Google AI Overview.
This track covers Claude Code workflows, WebMCP agent interfaces, context management, evidence gates, RAG, and the operational decisions that move an AI idea into production.
Best for
builders, founders, and engineers shipping with AI
Reading depth
20 implementation notes, case studies, and security-automation architectures
Start here
Claude Code Best Practices the Official Docs Don't Cover (2026)
What I learned building 36K lines of production code with Claude Code: quality gates, multi-agent orchestration, and the workflow patterns that ship.
This track reframes parallel thinking, novelty seeking, abstraction, and chaos tolerance as engineering leverage, then turns them into practical workflow scaffolding.
Best for
ADHD and neurodivergent developers, designers, and founders
Reading depth
7 core essays and workflow guides
Start here
ADHD Productivity: The System I Built After GTD Failed Me
GTD doesn't work for ADHD brains. The energy-aware productivity system I built instead — hyperfocus scheduling, AI processing, and the workflow I use to ship.
Cluster spotlights
AI Visibility Engineering
Featured entry
Answer Engine Optimization: 6 Factors That Decide If AI Cites You
Answer engine optimization (AEO) determines which sites AI search engines cite. The 6 factors driving citations in Perplexity, ChatGPT, and Google AI Overview.
Supporting reads
How to Get Perplexity and ChatGPT to Cite Your Website
Step-by-step guide to make your content visible in AI search engines. Includes robots.txt, structured data, and content format optimization.
llms.txt for AI Crawlers: Why robots.txt Is Not Enough
robots.txt was built for search engines, not AI crawlers. llms.txt gives LLMs structured context about your site that robots.txt can never provide.
Build with AI
Featured entry
Claude Code Best Practices the Official Docs Don't Cover (2026)
What I learned building 36K lines of production code with Claude Code: quality gates, multi-agent orchestration, and the workflow patterns that ship.
Supporting reads
Claude Was Eating My Tokens. This 3-Tier System Cut Usage 60%.
Progressive disclosure cuts AI token costs by 40%. Learn the 3-tier system that reduced my Claude expenses while improving output quality.
RAG Explained: How to Stop LLMs From Making Things Up
RAG retrieves live data to fix LLM hallucinations. Build accurate AI apps with up-to-date knowledge sources without retraining or fine-tuning models.
Think Better with ADHD
Featured entry
ADHD Productivity: The System I Built After GTD Failed Me
GTD doesn't work for ADHD brains. The energy-aware productivity system I built instead — hyperfocus scheduling, AI processing, and the workflow I use to ship.
Supporting reads
Reading paths
AI Visibility Engineering
Answer Engine Optimization: 6 Factors That Decide If AI Cites You
Answer engine optimization (AEO) determines which sites AI search engines cite. The 6 factors driving citations in Perplexity, ChatGPT, and Google AI Overview.
Build with AI
Claude Code Best Practices the Official Docs Don't Cover (2026)
What I learned building 36K lines of production code with Claude Code: quality gates, multi-agent orchestration, and the workflow patterns that ship.
Think Better with ADHD
ADHD Productivity: The System I Built After GTD Failed Me
GTD doesn't work for ADHD brains. The energy-aware productivity system I built instead — hyperfocus scheduling, AI processing, and the workflow I use to ship.
Product bridge
AI Visibility Auditing
AI visibility auditing platform. 10 verified infrastructure checks measuring whether AI systems can find, recommend, and cite your website.
Open Source Framework
Open-source framework for measuring whether AI systems can discover, retrieve, and cite website content. 15 checks across crawlability, structured data, and AI-specific signals.
Futures Trading System
CME Micro Bitcoin Futures (MBT) trading bot for Bulenox prop firm. Rithmic connectivity, fade strategy, quant factory for signal research.
Latest dispatches
In March 2026 I had 3 half-written posts and 0 shipped. By May 2026 the count was 63 published. The change was not effort. It was an externalized state machine that stopped using working memory as the index.
Two AI agents wrote decisions to the same Convex table for the first time on May 12, 2026. Four rows landed in seven hours. Here is the schema, the gate, and why it matters when you run more than one agent on the same work.
Bing AI cited my site 8 times a day after I stopped tuning individual pages. The principle: entity-level SEO is the floor; page-level work is the ceiling.
Under subscription plans, the cost of an LLM call is plan-window burn, not API tokens. That changes which decisions you tune and which work belongs off-LLM entirely. A manifesto for the 5-hour rate-limit era.