Augmented LLM

Definition

The basic building block of any agentic system: an LLM augmented with retrieval, tools, and memory, which the model actively uses to accomplish a task.

Key points

• The model actively uses these augmentations — it can generate its own search queries, select tools, and decide what to retain.
• Two recommendations: tailor the augmentations to the use case, and ensure they provide an easy, well-documented interface for the LLM.
• All workflow and agent patterns are composed from this one building block.

Why it matters for the exam

• "Augmented LLM" is the named foundation for the pattern taxonomy; retrieval + tools + memory is a testable triad.

Common gotchas

• Start simple; only add agentic complexity when it demonstrably improves outcomes — a single augmented LLM call often suffices.

See also

• Workflows vs Agents
• Autonomous Agents
• Agentic loop

Sources

• blog-building-effective-agents