AI Product Engineering
Block 3: Agents, Evaluation & Deployment
A language model only produces text. Tool calling lets it request execution of real functions — and use the results.
The model decides to call the tool. Your code executes it.
Pick the strongest answer.
Pick the strongest answer.
Pick the strongest answer.
name: 'calculate'
description: 'Evaluate a mathematical expression.'
parameters:
expression:
type: string
description: 'Expression to evaluate, e.g. "(15 / 100) * 2340"'
required: trueThe model reads the description to decide when to use this tool. Write it clearly.
eval() to AI-driven tool callsYour RAG retrieval function from Session 9 becomes a tool the agent can call.
User asks a question → agent decides to search knowledge base → retrieves relevant chunks → generates grounded answer.
Sessions 9 + 11 = a complete agentic knowledge assistant.
True or false — call on a pair.
True or false — call on a pair.
True or false — call on a pair.
POST /tools + queryWithTools — ask (15 / 100) * 2340/query — do not replace it with tools-onlysearch_knowledge_base in executeTool + register declarationprompts.mdAlso test: explain an API (no tool) · vault question (search tool).
Next Session: Multimodal AI & Data Pipelines