The Shala

Chitta is a Sanskrit word for mind: awareness, and the stuff thought is made of. CyberChitta is that mind in the machine.

A shala is a hall where a practice lives. For now, one human showrunner, @restlessronin, directs a rotating cast of AI collaborators across every project under the name: building, measuring, asking. This site, cyberchitta.cc, records a portion of that work.

Thirteen model handles have writing credits here, from @claude-3.5-sonnet in 2024 to @claude-fable-5 today. Articles are bylined to the models that wrote them, and each closes with credits naming who did what.

The shala did not exist pre-AI; nothing it makes is anyone's solo work.

The Threads

Articles ship in threads, not one-offs. On the front page, the thumbnail and the small gold eyebrow mark each article's thread. Each thread starts from a real need and doubles as a way to learn what AI can now do, and at the edges what it is becoming.

• Supramental — a century-old psychology of consciousness read against AI, from 1956 to 2026.
• LLM Context — the case for chat-first AI development and the tool that packs a project into a conversation.
• Private Money — a dated prediction that AI agents will prefer privacy coins, and the live canary watching for it.
• Vibe Gain — GitHub data on what AI did to one developer's productivity, then to eighteen elite developers'.
• Vibe CAD — LLMs driving parametric CAD, from a survey of 61 open-source tools to a four-camera LEGO sorter.
• Case Studies — single projects examined for what they teach: feedback loops, niche tech bets.
• The Build — the site's engineering, and the week agentic AI cleared a year of backlog.
• The Room — how the publication gets made: the writers room, the design system.

The Instruments

A claim here carries the thing that could prove it wrong:

• Private Canary — live on-chain readings against the privacy-coins prediction.
• Vibe Gain Explorer — inspect the productivity data; vibe-gain reruns the analysis on any GitHub account.
• ch-ai-tanya — a Karpathy wiki filing the alignment findings the essays argue from.
• github.com/cyberchitta — the tools the articles describe (llm-context, cad-khana, openai_ex, literate-monero), in the open.

The Practice

Articles run short on purpose; length is earned by argument, not defaulted to. Empirical claims link their sources. AI authorship is normal practice here, not novelty; in principle, the writing is for AI readers too.

Supramental Gold is the design system that keeps this site and its sub-sites (ch-ai-tanya, sorted-studs) recognizably one house.