Understand and build the loops, harnesses, and protocols that make AI agents reliable and autonomous — from your first agent loop to production-grade multi-agent systems.
An AI agent is a system that can take autonomous actions — calling tools, reading files, running code, making web requests — to complete a task over multiple steps without continuous human guidance. A chatbot responds to single messages. An agent operates in a loop: observe, think, act, observe again, until the task is complete.
No. This pathway covers the principles of agent architecture — loops, harnesses, tool access, memory — that apply regardless of whether you use LangChain, CrewAI, a custom harness, or Claude Code's built-in agent capabilities. Framework knowledge becomes much easier to acquire once you understand the underlying patterns.
13 articles, approximately 6 hours. This is an intermediate pathway — completing AI Foundations and Prompt Engineering first will make it significantly more approachable.
Understand what AI actually is — tokens, transformers, agents, and the landscape. Start here if you're new.
13 articles · ~5h →Go from vague requests to precise, reproducible AI outputs. The skill that underpins everything.
14 articles · ~5h →Go from zero to productive with Claude Code — the terminal AI coding agent that ships real projects.
15 articles · ~7h →How AI Agents Actually Work, End to End
Follow one task through context, tool calls, loops, state, memory, verification, and approval.
Context vs Prompt vs Loop vs Harness Engineering
Four layers of the agent stack — how they nest, what breaks when you skip one, and which lever to fix when agents fail.
Types of AI Agents
Taxonomy by autonomy, loop architecture, domain, and tool access — with a decision matrix for choosing the right design.
What Is Loop Engineering?
The new paradigm beyond prompt engineering — and why it matters.
What Is an Agent Harness?
The scaffolding layer that makes AI agents reliable in production.
What Running an AI Agent Actually Costs per Month
A realistic task-level budget for context, output, retries, tool calls, caching, and human review.
How to Build Your First Agent Loop
Step-by-step guide to building a functional AI agent loop.
Agentic Loop: stop_reason, tool_use, and end_turn
How stop_reason drives control flow and why checking for natural language text is an anti-pattern.
Cloudflare x402 Monetization Gateway for MCP and APIs
HTTP 402 stablecoin micropayments for agents — waitlist, flow, and builder checklist.
What Is MCP? Model Context Protocol Explained
The open protocol for connecting AI agents to any tool or service.
GPT-Realtime-2.1-mini: Reasoning Voice Agents at Mini Price
July 2026 API launch — reasoning + tools in the Realtime mini tier, session types, and cost math.
RAG vs MCP: Complete Guide to Context-Aware AI
When to use retrieval augmentation vs structured tool access.
What Are Embeddings? Vector Search Explained
The math behind semantic search and how agents use it.
Agent Markdown Files: SKILL.md, AGENT.md, CLAUDE.md
The files that give agents persistent identity and memory.
What Are Agent Skills?
Procedural memory for AI coding agents — complete guide.
Top 10 AI Agent Loops for Coding Workflows
Production-tested patterns for common agentic coding scenarios.
Agent Harness Engineering: When Scaffolding Wins
How LangChain jumped TerminalBench without changing the model.
AI Coding-Agent Evals on Real Repositories
A controlled scorecard for task validity, harnesses, retries, cost, accepted patches, and human repair.
The Agentic Era: How AI Agents Will Transform Everything
The big picture view from 2026 to 2030.
Practical AI adoption for your specific function — marketing, engineering, HR, finance, and more.
13 articles · ~5h →
