Curriculum
From day one to certified
Seven levels of practitioner-led training that take your engineering team from "what is an LLM" through production-grade multi-agent systems. Every technique on the current frontier is covered somewhere in this tree. Refreshed quarterly.
The progression
Every learner path follows the same ladder. Levels build on each other — skip what you already know, but don't skip what you think you know.
Literacy
4-6 hours · Everyone, technical or not
What LLMs are and aren't, what never to paste, how to spot a hallucination, honest expectations. The foundation every person in your org needs before they touch a tool.
Practitioner Basics
6-8 hours · Engineers and technical contributors
Prompting patterns, picking the right mode, reading AI diffs, small verifiable steps, context hygiene, when not to use AI.
Prompt Engineering
5-7 hours · Anyone writing prompts for real work
Zero/few-shot, chain-of-thought, tree-of-thought, ReAct, structured outputs, prompt chaining, reflection, and evals.
Tool Fluency
10-14 hours per tool · Daily users of Claude Code or Codex/Copilot
CLAUDE.md, the agent loop, MCPs, subagents, plan mode, skills, cost discipline, and debugging when things go sideways. Separate tracks for Claude Code and Codex/Copilot.
Specialist Tracks
6-10 hours per track · Backend, data, DevOps, security, SRE engineers
Discipline-specific playbooks: the traps your specialty has that a generalist would miss. Pick one, unlock others later.
Leadership & Governance
8-12 hours · VP Eng, CTO, CISO, Head of L&D
Rollout strategy, measuring adoption honestly, policy and governance, risk registers, talent impact. The course your leadership team needs before they make decisions about AI.
Advanced Techniques
16-24 hours · Senior engineers and AI platform builders
Context engineering, multi-agent orchestration, RAG that works, MCP authoring, production evals, guardrails, cost optimization, and observability. The frontier.
5 Foundations Tracks
The customer-facing bundles. Each track pulls lessons from multiple levels and reorders them into a journey that makes sense for that audience. Pick the one that fits your team, or bundle several for full-org coverage.
Claude Code for Working Developers
Ship real PRs with an AI partner
Codex & Copilot Mastery
Inline completion done right
AI for Engineering Managers
Rollout, measurement, policy, risk
AI Security & Policy
Threat models, not theater
Agents & MCP Servers
Build what the frontier looks like
How each track works
Atomic lessons
5-10 minute videos, each followed by a hands-on sandbox exercise. No lesson has more than 15 minutes of passive content before you do something.
Real repos, not toy examples
Labs use realistic codebases with tests, lint, CI, and deploy pipelines. Auto-graded where possible, human-reviewed where it matters.
4-tier certification
Bronze, Silver, Gold, and Certified. Each tier is earned through real work, computed server-side from progress. Exportable as Open Badges 2.0.
Not sure where to start?
Take our free 3-minute AI maturity assessment. Score your organization on 8 dimensions and get a personalized 90-day action plan that tells you exactly which tracks to start with.