Agents & Retrieval
Understand tool-using assistants, document search, retrieval and agent failures.
Choose a lesson for its Quick Review. The questions and answers are being prepared.
Agent and Tool Fundamentals
Agents Versus Fixed Workflows
Distinguish planned automation steps from model-guided tool use.
Tools, Functions and Permissions
Read a tool description, expected inputs and access boundaries.
Tool Selection and Execution Order
Follow why an agent chooses a tool and the order of actions.
Human Approval and Safe Stop Conditions
Design when an agent must stop, ask for confirmation or hand off.
Knowledge and Retrieval
Search and Retrieval Fundamentals
Understand how external information becomes available to an AI system.
Document Chunking and Metadata
Explore splitting documents and keeping source information attached.
Embeddings and Similarity Search
Understand conceptual vector similarity without requiring real vector APIs.
RAG: Retrieve, Ground and Answer
Trace retrieval-augmented generation and evaluate source alignment.
Agent Reliability and Diagnostics
Missing, Stale and Irrelevant Retrieval
Identify when retrieval misses or uses the wrong information.
Tool Failure Versus Model Failure
Separate an invalid tool call from a poor generated explanation.
Untrusted Documents and Prompt Injection
Identify instructions hidden inside retrieved content and safe boundaries.
Agent Traces and Recovery Paths
Follow tool-call history, error states, retries and safe fallbacks.