What are AI Agents

Lessons in this group, roughly in build order:

• what-are-ai-agents — An AI agent is an LLM placed inside a control loop that lets it observe state, decide on actions, call…
• agent-loop — The agent loop is the driver that repeatedly calls the model, executes any tool it requests, appends the…
• perception-user-input — Perception is how an agent ingests its world each turn — the user request plus any retrieved context, tool…
• reason-and-plan — The reasoning step is where the model, given the current context, decides what to do next — break down the…
• acting-tool-invocation — Acting is the step where the agent leaves the model and touches the world — calling a function, query, or…
• observation-reflection — Observation is taking a tool’s result back into context; reflection is the model judging that result — did…