Lesson 5: Tool Use & Execution

Advanced Architectures for Vertical AI Agents ~3 min read 60 min hands-on

Tool and Retrieval Integration

Objective

Build and operate a self-contained platform that combines:

  • LLM function calling

  • schema-validated tool execution

  • iterative execution loop orchestration

  • naive retrieval-augmented generation over local corpus data

Delivered System

  • Child project: tool-rag-orchestration-platform

  • Runtime surfaces:

  • API backend

  • dashboard frontend

  • containerized deployment

Core Capabilities

  • Function calling interface for external actions

  • Pydantic-based schema validation for tool inputs and outputs

  • Tool executor with safe invocation and execution logs

  • Corpus ingestion, chunking, indexing, and retrieval-backed responses

Operations

  • Run from child directory scripts:

  • ./scripts/build.sh

  • ./scripts/start.sh

  • ./scripts/demo.sh

  • ./scripts/test.sh

  • ./scripts/cleanup.sh

Notes

  • Child project is self-contained and can be used directly without external wrapper scripts.

  • Sensitive values should remain empty in .env unless explicitly provided for local runtime.

System Architecture

Keep going with the full lesson

The rest of this lesson — plus its implementation guide, runnable scripts and source code — is included with the course or an All Access plan. Create a free account to start the free lessons in every course.

Unlock the full lesson Create free account

Already a member? Sign in