Chapter overviews

Setting Up Your Agent Development Workspace

Connect to major LLM providers.

Design platform architecture with self-service portal, service catalog, and golden paths.

Implementing Multi-Layer Defense Against Direct and Indirect Prompt Injection

Build automated use case scoring and data readiness evaluation tools for client discovery workshops.

Implement branching strategies, code review workflows, and merge policies for AI projects.

Build FastAPI streaming chat endpoints using SSE for token-by-token delivery across four LLM providers.

Build an automated ADR (Architecture Decision Record) system that captures, validates, and enforces GenAI-specific technology decisions across model selection, hosting strategy, and RAG-vs-fine-tuning trade-offs.

Build document ingestion pipelines using Docling, Google Document AI, and vision-language models for PDF, HTML, and DOCX extraction with structure preservation.

Design interview loops covering coding, system design, LLM-specific, and evaluation challenges.

Design team topologies including platform team, agent team, ML team, and hybrid structures.

Build career frameworks with IC track early-career-distinguished, management track, and specialist tracks.

Curate, version, and maintain evaluation datasets for hosted LLMs.

Implement Technical Requirements for EU AI Act Risk Categories with Current Enforcement Timeline

Understanding API-Based Inference Systems

Build CI/CD pipelines using Argo Workflows for prompts, model configs, RAG configs, and guardrail policies.

Build a Docker container for a Python app that calls the Gemini API, then run it locally and push to a registry.

Processing Data Streams Efficiently

Write reusable, modular code.

Build REST APIs with FastAPI, Pydantic validation, and OpenAPI docs.