The New AI Stack
Understanding Models, Agents, and Skills as the new computing paradigm
Showing 13-24 of 56
Addy Osmani's 5 Building Blocks of Loops
Addy Osmani frames self-improving agents as a handful of building blocks plus memory. Here is how that loop-engineering framework maps to real agent loops you can build.
Exit Conditions: The Heart of Every Loop
The exit condition, plus a max-iterations cap, check command, and anti-gaming guardrails, is what separates a useful agent loop from an infinite one.
Loop Engineering vs Prompt Engineering
A head-to-head comparison of loop engineering and prompt engineering: inputs, failure modes, cost, and exactly when to use each in 2026.
Runtime Hacking for AI Tool Builders
Advanced runtime techniques for AI skill development. Learn hot reloading, dynamic evaluation, monkey patching, and environment manipulation for faster iteration.
The Convergence: OpenClaw, Claude Code, and Skills
OpenClaw and Claude Code are converging toward a unified skill layer. Markdown config, skill registries, tool use, memory -- the architecture is aligning.
Surprising Behaviors in AI Extensions
Discover the edge cases that emerge when AI modifies existing code. Learn patterns to predict, prevent, and harness unexpected behaviors in AI-powered extensions.
What Is Loop Engineering? The 2026 Meta
Loop engineering is the 2026 successor to prompt engineering: designing autonomous agent loops that act, observe, decide, and repeat until a verifiable exit condition.
Modern Objective-C Meets AI
How AI coding assistants handle legacy Objective-C codebases. Practical strategies for using AI to maintain, modernize, and extend Objective-C applications.
Thread Safety in AI Applications
Concurrency patterns that AI tools must respect when generating multi-threaded code. Practical patterns for thread-safe AI-generated applications.
Real-Time Collaboration With AI
Multi-user AI development environments where teams build together with shared AI assistants. Architecture patterns and emerging tools for real-time AI collaboration.
AI Agent vs Chatbot: Why the Distinction Matters
Agents use tools, persist state, and take autonomous action. Chatbots just respond. Understanding this distinction explains why skill ecosystems exist.
The Case for Killing Boilerplate
AI makes old code patterns obsolete. Why boilerplate code, scaffolding, and repetitive patterns should be generated on demand rather than maintained by hand.