Curriculum
What your team will learn
Six core modules covering everything from foundational AI thinking through agent architecture and workflow design. Sessions are mixed and matched to build the right program for your team — no module is mandatory, and every session is tailored to your specific context.
AI Fundamentals & Mental Models
The right way to think about what large language models are, what they're not, how they reason, and why they fail. Builds the foundation for everything else.
Topics covered
- How LLMs actually work (non-technical)
- Probabilistic reasoning vs. deterministic logic
- Where AI excels vs. where it hallucinates
- How to evaluate AI output critically
- Mental models for working with AI as a collaborator
Prompt Engineering
The craft of writing prompts that produce reliable, repeatable results. Covers the full toolkit from basic formatting through advanced techniques.
Topics covered
- System prompts: structure, tone, constraints
- Role assignment and persona setting
- Chain-of-thought and step-by-step reasoning
- Few-shot examples and in-context learning
- Output format control (JSON, markdown, structured data)
- Prompt debugging and iteration
AI Agents: Architecture & Use Cases
What an agent actually is, how tool use works, when a multi-agent architecture makes sense, and how to evaluate agent behavior in production.
Topics covered
- Agents vs. chat interfaces: the real difference
- Tool use: web search, code execution, file access, API calls
- Orchestrator and sub-agent patterns
- Multi-agent systems: when and why
- Human-in-the-loop design
- Evaluating and debugging agent behavior
Claude In Depth
Claude-specific capabilities, context window behavior, memory patterns, and how to design systems that take full advantage of what Claude does best.
Topics covered
- Context window mechanics and long-document handling
- Memory: in-context, external storage, RAG patterns
- Claude's constitutional approach and how it affects prompting
- Extended thinking mode: when and how to use it
- Claude API: tools, system prompts, conversation structure
- Comparing Claude to GPT-4 and Gemini for specific use cases
AI Workflow Design
How to design workflows where AI handles the right tasks and humans handle the right tasks — with clear routing logic, oversight, and failure modes.
Topics covered
- Mapping workflows for AI integration
- Task decomposition for AI
- Routing logic: when to escalate to human review
- Document-heavy workflows (contracts, reports, forms)
- Communication workflows (email, summaries, drafting)
- Measuring and validating AI workflow performance
Model Selection & Strategy
A practical framework for evaluating which AI tools fit which jobs — and how to build an AI strategy that doesn't lock you into one vendor or one tool.
Topics covered
- Claude vs. Gemini vs. GPT: capability comparison
- Evaluating models for specific task types
- Build vs. buy: custom tooling vs. off-the-shelf
- Vendor lock-in risk and mitigation
- AI governance and oversight at the organizational level
- Staying current as models evolve
Build the right program for your team
No module is delivered off the shelf. We start with a discovery call, I learn what your team actually needs, and we design a curriculum from there.
Book a discovery call