A technical, practical path to becoming an AI transformation consultant for small and mid-size companies, built on Anthropic's Claude products. Every module ends with a lab and a deliverable.
Read the full syllabus · Source on GitHub
| 1 | Transformation landscape Levels of AI change, model choice, Claude products, privacy, evidence |
| 2 | Prompt and context engineering Prompt patterns, structured output, tested prompt library |
| 3 | Claude apps for knowledge work Projects, connectors, scheduled tasks, rollout playbook |
| 4 | Claude API in Python Tool use, caching, batches, invoice extractor |
| 5 | Knowledge retrieval (RAG) Hebrew/English document assistant with citations |
| 6 | Model Context Protocol (MCP) Build and connect MCP servers |
| 7 | Claude Code as a delivery engine CLAUDE.md, hooks, subagents, project template |
| 8 | Agents, Skills and plugins Agent patterns, Agent SDK, Skills, lead-to-proposal agent |
| 9 | Evaluation, cost and reliability Eval sets, graders, routing, monitoring |
| 10 | Security, governance and compliance Prompt injection, PII, Israeli privacy law, policy templates |
| 11 | Discovery, ROI and proposals Readiness audit, scoring, ROI model, SOW |
| 12 | Capstone and positioning Case study, positioning, 90-day go-to-market |
Written October 2026. Check model names, API parameters and Claude Code settings against docs.claude.com before relying on them. Lab code has not yet been run against the live API. Module 10 is not legal advice.