Explain agentic AI architecture.

Agentic AI architecture refers to systems that act autonomously to achieve goals. Core components include:

1. Perception – sensors or data ingestion modules that interpret the environment.
2. Knowledge Base – structured representation of facts, rules, or learned models.
3. Planning – algorithms (e.g., hierarchical RL, symbolic planners) that generate action sequences.
4. Decision‑Making – policy networks or rule engines that select actions.
5. Execution – actuators or API calls that perform actions.
6. Learning – mechanisms to update the knowledge base or policy based on feedback.

Typical examples are autonomous robots, virtual assistants, or self‑optimizing data pipelines. The architecture emphasizes modularity, feedback loops, and the ability to adapt to new goals or constraints.