Archive Interface
The archive is structured for quick search, category filtering, and fast scanning without the visual noise of a generic blog grid.
Archive Snapshot
Search the whole library or jump into a category without losing the editorial structure.
Discover Git's object model. Learn how blobs, trees, and commits form a directed acyclic graph, and how the index enables fast staging and diffing operations.
Git Isn't a Diff Tracker: How Blobs, Trees, DAG Commits, and the Index Actually Work Under the Hood
Build a fully functional agentic AI that runs entirely on-device at 142KB. Complete guide to model quantization, tiny agent frameworks, and mobile deployment for offline-first LLM applications.
How I Shrank an Agentic AI to 142KB and Ran It on a $150 Phone — A Practical Guide to Local-First LLM Agents
Explore how Rust-powered clean-room imaging pipelines and offline AI agents enable truly local-first applications, with architecture patterns, code examples, and performance insights.
From Rust to Reality: How Clean-Room Photography Engines and Offline AI Agents Are Redefining Local-First Software
A deep-dive into screen-free AI architecture: offline models, voice-only interfaces, and walk-driven storytelling that gets developers moving.
The Touch Grass Manifesto: Building AI Apps That Force Developers Outside — Offline Models, Voice-Only Interfaces, and Walk-Driven Storytelling Without a Single Pixel on Screen
Explore how local-first AI architectures enable a 142KB offline vision mentor to run on low-end hardware, eliminating cloud dependencies and latency for on-device inference.
The Rise of Local-First AI: Building a 142KB Offline Vision Mentor on a $150 Phone
How offline-first, voice-first, and real-world-aware AI apps are disrupting cloud-heavy developer tooling — with architecture patterns, trade-offs, and production implementation strategies.
The 'Touch Grass' Movement in Dev Tools: Why Offline, Voice-First, and Real-World-Aware AI Apps Are Beating Cloud-Heavy Monoliths
Explore the architecture and engineering principles behind autonomous AI agents that self-improve through internal simulation, self-play, and curiosity-driven exploration—getting stronger without human interaction.
The Touch Grass Paradox: Engineering AI Agents That Get Stronger the Less You Use Them
Explore how agent-ready design systems bridge human usability and machine comprehension, reshaping frontend architecture with semantic APIs, structured tokens, and dual-consumption patterns.
Designing for Both Humans and Machines: How 'Agent-Ready' Design Systems Are Reshaping the Frontend Stack
AI agent frameworks are exploding on GitHub, but the fundamentals—merge conflict resolution, version control, code ownership—are being ignored. Here's why that's dangerous.
60K Stars for AI Agent Frameworks, Zero Stars for Merge Conflicts: Why the Hottest Developer Trend Is Actually Going Back to Basics
An experienced developer breaks down which coding tasks AI genuinely replaced, which it merely shortcut, and why the distinction matters for your career and code quality.
AI Got Better While I Was Away — But Did We Get Dumber? A Developer's Honest Audit of What AI Actually Replaced vs. What It Just Made Lazy
Discover the engineering patterns to slash AI code completion costs by 80%. Learn context pruning, model cascading, and caching strategies for GPT-4o and Claude.
Stop Burning $50/Day on AI Code Completions: A Senior Engineer's Token Optimization Playbook for Frontier Models in 2025
Discover how single-prompt bloat inflates API costs in AI coding. Learn to implement gascity-style multi-agent orchestration to optimize token usage and improve code quality.
Multi-Agent Orchestration vs. Single-Prompt Bloat: Why Your AI Coding Workflow Is Burning Tokens (and How to Fix It with gascity-style SDKs)