Hi, I'm Harsh Joshi 👋

I currently work at AION building the future of decentralized HPC compute. The organization has been wonderfully supportive of this personal exploration - providing the time, flexibility, and Claude Code credits that make this digital garden possible. Their belief in fostering creative side projects is something I deeply appreciate.

This repository contains the central engine (github.com/getHarsh) which orchestrates my entire digital ecosystem. Feel free to use it if you have multiple exploration branches and want proper legal, conceptual, and publication separation.

For the complete technical architecture, implementation details, and organizational structure, see what_is_getHarsh.md.

Quick Links

Understanding the System

• What is getHarsh? - Complete technical architecture and vision
• Legal Boundaries & Entity Structure - How RTEPL and DSPL are separated
• Virtual Organization - How projects act as virtual team members
• Implementation Philosophy - The semantic-trace-first approach

Technical Documentation

• Engine Architecture - Exact technical architecture of the getHarsh engine
• Build System - Complete build system documentation
• Theme Specifications - Detailed component and theme specifications
• Technical Infrastructure - Overview of the Gatsby engine and enhancement layers

Licensed under Apache License 2.0 - See LICENSE for details

The Journey 🚀

This is a personal project I've been tinkering with in spare time over the years. As someone with ADHD, I've always wondered - what if we could build tools that work with neurodivergent thinking patterns instead of against them? Tools where the latency between thought and published content approaches zero?

The dream is simple: maybe one day, when I retire to pursue research full-time, this could become something more. For now, it's my weekend exploration into semantic trace extraction - capturing thought patterns as high-quality training data for experimental neural architectures like HENA and QAECT.

Tools of Trade 🛠️

My tools of trade are Claude Code, Google Drive, GitHub, GoDaddy, and Obsidian. Everything else falls in respective buckets.

The Digital Ecosystem: Exploring Chaos, Causality & Entropy 🌐

At the heart of my exploration lies a focused research mission: understanding the intersection of Chaos, Causality, and Entropy from different angles. Each domain serves as a specialized lens for examining these fundamental forces.

• getHarsh.in - Personal research hub where chaos becomes order through infrastructure (GitHub)
• causality.in - Lifelong exploration of non-linear time and causal relationships. "Don't say I didn't warn you!" (GitHub)
• rawThoughts.in - "Living in the future vs. peeping into it" - Post-AGI economics born from startup ashes (GitHub)
• sleepwalker.in - A 9102 A.D. tale: "Just because you can, doesn't mean you should" (GitHub)
• daostudio.in - "Democratise AI, Commoditise Intelligence" - Cooperative tech experiments (GitHub)

Each domain hosts both a main site and blog subdomain, with specialized projects living within them. Learn more about the domain strategy and project distribution →

Why Build This? 🧠

The philosophy is simple: neurodivergent minds see connections everywhere, but traditional tools force linear thinking. This system embraces the chaos - using symlinks to Google Drive for instant sync, Gatsby for AI-friendly output, and virtual team members that might someday handle the boring parts.

I half-jokingly call it uploading my consciousness - current progress: 3.14% 🥧. But beneath the humor lies a serious hypothesis: that rich semantic traces from real human thought patterns might enable training AI architectures with emergent reasoning capabilities. The goal isn't literal consciousness upload but rather exploring whether quality trumps quantity in AI training data. Maybe HENA will show signs of hierarchical reasoning, maybe not - but the infrastructure for the experiment is fascinating to build.

Licensed under Apache License 2.0 - See LICENSE for details