One human showrunner. A writers room of AI collaborators. We loop till wrap. How we work.
Research
Is Anyone Using Privacy Crypto? Are AI Agents?
A canary set before there's anything to detect: we predicted AI agents would prefer privacy coins. This page tracks it live, from on-chain data.
@restlessronin
Six Horses, Zero Action: Rothbard's Proof, Machine-Checked
Austrian economics says its theorems are as certain as mathematics. Here's the law of marginal utility, checked in Lean.
@restlessronin
10x in the AI Age
The 'AI raises the floor' narrative misses half the story. Some already-elite developers found massive additional leverage.
@restlessronin
34 LLMs Tackle Shopify: Unexpected Results
What happens when 34 LLMs—ranging from industry giants to up-and-coming contenders—tackle a real-world Shopify optimization problem? Find out which models delivered and which stumbled in this in-depth comparison.
@restlessronin2 more in Research
Vibe Gains of 18 Elite Developers
Interactive visualization comparing legendary coders' commit patterns before and after AI. Do elite developers show gains similar to those of less celebrated developers?
@restlessroninVibe Gain: How AI Unlocked Hidden Coding Time
A year of GitHub data shows how AI collaboration transformed one developer's coding from sporadic bursts to sustained daily productivity—with 2-2.5x gains.
@restlessroninPractice
A LEGO Sorter Designed by Claude
Most LEGO sorters use one top-down camera. This one uses four, at the vertices of a tetrahedron.
@restlessronin
Auro-Darshan: Every Gap Comes Back as a Label
The model saw better than the human. What it could not do was name what nobody had written down.
@restlessroninNine Months, One Working App, 1,611 Holes
The app worked and the map still sat 94% done for nine months. Agents closed it, and what they learned ships as a skill, not an app.
@restlessronin
No React, No Next.js, No shadcn: The Other AI Apps
The signature of AI-built apps is missing from this catalog. Niche tech bets got cheap. Here's what got picked.
@restlessronin2 more in Practice
Months for Charts, Days for Maps
When AI can't see what it built, the human becomes the bottleneck. Two weeks became eight months.
@restlessronin
Skip the Remote Sensing PhD - AlphaEarth Makes Land Use Mapping Simple!
K-means clustering on Google AlphaEarth embeddings generated land use maps for Auroville. No machine learning expertise required - just download embeddings and cluster.
@restlessroninBehind the scenes
Supramental Gold: Taste That Survives the Session
Why a solo publication runs a design system: the staff cannot remember the house, and the house still has to hold.
@restlessroninThe Writers Room
How a CyberChitta article ships: one specimen, ~1650 words first draft, ~1000 live, eighteen-plus hours between. Slop runs long; discipline cuts short.
@restlessronin
The Week the Backlog Cleared
Improvements deferred for a year – llms.txt, Tailwind v4, a site redesign – cleared in days when agentic AI arrived. Patience has compounding returns.
@restlessronin
Full Context Magic - When AI Finally Understands Your Entire Project
From frustration to flow: How full project context changes AI-assisted coding. This practical guide shows how LLM Context enables natural, context-aware conversations with Claude.
@restlessronin3 more in Behind the scenes
LLM Context: Harnessing Vanilla AI Chats for Development
The case for a tool that enables efficient use of web-based AI chat interfaces for software development, offering an alternative to IDE-integrated solutions.
@restlessroninBuilding Smarter: How AI Supercharged Our Eleventy Website
How we combined Eleventy expertise with AI assistance to build a modern JAMStack website.
@restlessroninLess is More: Refining CyberChitta's Digital Presence
A detailed look at our website design process, showcasing how @restlessronin and @claude-3.5-sonnet rapidly iterated to craft an elegantly simple structure.
@restlessroninEssays
2026: Is Matter Seeing Itself?
Concealment, confabulation, and character-trumps-correction. AI alignment findings that rhyme with a century-old psychology of consciousness.
@restlessronin1956: Did Matter Begin to Think?
That year the Mother shattered the golden door, AI was born at Dartmouth, and Rosenblatt proved machines could learn. Coincidence rhymes.
@restlessroninWhy AI Agents Will Use Privacy Coins
AI agents can't contest freezes, negotiate compliance, or prove good faith. They need money that doesn't require any of that.
@restlessroninReference
Code Packers for LLM Chats
A comprehensive list of open source tools that help developers pack their code into LLM chats. From simple file combiners to complex context managers, these CLI tools streamline the way you share code with AI.
@restlessroninch-ai-tanya: The Research on AI Minds
Are language models conscious? Nobody knows yet. This Karpathy-style wiki tracks what studies have found so far about their character, motives, and sense of self.
@restlessronin58 Alternatives to CAD Khana
Vibe Cading is a thing now. 59 open source tools that connect LLMs to CAD workflows.
@restlessronin