Buildready5/10
AI Agents
Planning, state machines, orchestration, handoffs, evaluation, security, human oversight, browser and computer-use tools.
5materials
41min reading
2levels
Start2
Core2
Build2
Advanced4
Learning route
Study route
Core level
3- 01Claude Code subscription vs API: choosing access and billingA practical comparison of Claude Code through Pro, Max, Team, or Enterprise and token-based access through Anthropic Console or a cloud provider across billing, identity, limits, automation, observability, and exit readiness.Core level9 min
- 02Codex subscription vs API: choosing access, billing, and automationA practical comparison of Codex through a ChatGPT plan versus your own OpenAI API key across billing, identity, limits, local and cloud tasks, CI, governance, observability, and migration.Core level10 min
- 03Human-in-the-loop for AIA practical production architecture for human oversight: involve a person at a specific risk point with enough context for real, not ceremonial, control. Covers contracts, authority boundaries, failure modes, evaluation, and controlled rollout.Core level6 min
Advanced level
2- 01AI agent harness: how to design a reliable runtimeA practical guide to AI agent harnesses: execution loops, tools, sandboxes, durable state, context assembly, permissions, checkpoints, evals, observability, and recovery for long-running tasks.Advanced level8 min
- 02OpenAI Responses vs Claude Messages vs Gemini Interactions APIA practical comparison of the main OpenAI, Anthropic, and Google APIs for production AI: state, tools, streaming, background jobs, portability, evaluation, and migration controls.Advanced level8 min