Forty-six lessons. Nine schools. One ladder.

The ladder

Five levels. Each one ends in a proof a peer can check, and each is built from the lessons above.

1. Level 1 of 5

   Literate

   Use AI deliberately and know when it is wrong.

   ProofA documented ten-task AI-assisted workflow with an error log.

2. Level 2 of 5

   Fluent

   Wire models to tools and measure the result.

   ProofA tool-using assistant that passes a 20-case eval set at a threshold you set in advance.

3. Level 3 of 5

   Builder

   Run an agent for real users.

   ProofAn agent serving real users for 30 days, with traces and an incident record.

4. Level 4 of 5

   Orchestrator

   Coordinate many agents and humans, and prove it beats a baseline.

   ProofA gated multi-agent system that beats a baseline on a pre-registered eval, reproduced by a peer. You mentor one Builder.

5. Level 5 of 5

   Steward

   Set the standards others adopt.

   ProofA standard adopted by two outside teams, plus an audit of another team's system that finds and fixes real faults.

Swipe to see all five levels.

Browse the schools

Filter by school or format, search by keyword, and open any lesson for its prerequisites, what it unlocks and where it appears.

All46Literacy5Tools5Agents5Evals5Domain5Taste5Method5Civic5Studio6

All formatsLabSeminarStudioDuelField work

Showing 46 of 46 lessons

S1 · 5 lessons · 17 h

Model Literacy

How models work, fail and leak, so you trust them exactly as much as they deserve.

• 1.1LabHow models failPredict and log failures before they cost you.4 hFailure log
• 1.2StudioPrompts as specsWrite instructions with no room for misreading.3 hSpec
• 1.3LabContext budgetingSpend the context window like money.4 hContext audit
• 1.4SeminarReading benchmarksSeparate signal from leaderboard noise.3 hBenchmark memo
• 1.5SeminarData and privacyHandle client and personal data safely.3 hData policy

S2 · 5 lessons · 23 h

Tools and Context

Connect models to the world: functions, protocols, retrieval and memory.

• 2.1LabFunction callingWire a model to a tool that does something.4 hWorking tool call
• 2.2LabMCP serversExpose capabilities to any agent through a standard protocol.6 hMCP server
• 2.3StudioTool error designMake failures recoverable by the agent itself.4 hError catalogue
• 2.4LabRetrievalGround answers in sources you can cite.6 hRetrieval pipeline
• 2.5SeminarMemory systemsPersist what matters without drift.3 hMemory note

S3 · 5 lessons · 23 h

Agent Orchestration

Loops, handoffs, gates and permissions: how many agents do real work safely.

• 3.1LabAgent loop anatomyTrace one agent from goal to result.4 hAnnotated trace
• 3.2StudioMulti-agent handoffsSplit work without losing context.6 hHandoff protocol
• 3.3LabDecompositionBreak a goal into steps you can check.4 hTask graph
• 3.4StudioHuman gatesPlace approvals where the risk actually sits.4 hGate mapFrom our research
• 3.5LabPermissionsLimit what an agent can touch.5 hPermission policy

S4 · 5 lessons · 25 h

Evals and Verification

The skill employers rank first: knowing whether the output is good.

• 4.1LabDeterministic checksTest everything code can test.4 hTest suite
• 4.2LabJudge calibrationMeasure whether your automated judge deserves trust.6 hCalibration report
• 4.3StudioRubric designSay what good means, explicitly.4 hRubric
• 4.4LabTrace miningTurn logs into test cases.5 hCase set
• 4.5DuelEval duelRed team against blue team on a live system.6 hBreak report

S5 · 5 lessons · 24 h

Domain Judgment

Frame problems worth solving and price the cost of being wrong.

• 5.1SeminarProblem framingChoose a problem worth solving.3 hProblem brief
• 5.2Field workProcess mappingMap where the work really happens.8 hProcess map
• 5.3SeminarRisk and liabilityPrice the cost of being wrong.3 hRisk register
• 5.4Field workExpert interviewsExtract the knowledge nobody wrote down.6 hKnowledge base
• 5.5SeminarDecisions under uncertaintyCompare options with honest ranges.4 hDecision memoFrom our research

S6 · 5 lessons · 18 h

Taste and Expression

Critique, narrative and ethics: the human side of making things people trust.

• 6.1StudioCritique as methodJudge work by explicit reasons.4 hCritique log
• 6.2StudioWriting for decisionsMake your case in one page.4 hDecision brief
• 6.3SeminarNarrativeStructure a story people act on.3 hNarrative draft
• 6.4StudioCritiquing AI outputFind what is bland, wrong or unsafe.4 hAnnotated output
• 6.5SeminarEthics in artefactsTest your work against harm before it ships.3 hEthics reviewFrom our research

S7 · 5 lessons · 25 h

Science and Method

Design experiments that can fail, pre-register, replicate and forecast honestly.

• 7.1LabExperiment designDesign a test that could fail.6 hPre-registration draft
• 7.2SeminarPre-registrationCommit to metrics before you look.2 hRegistered plan
• 7.3LabReplicationReproduce someone else's result from their artefact alone.8 hReplication reportFrom our research
• 7.4DuelForecastingScore your own predictions and learn your calibration.4 hBrier score
• 7.5SeminarCausal inferenceSeparate causes from correlations.5 hCausal map

S8 · 5 lessons · 58 h

Governance and Civic

Policy, security, labour and public-sector deployment under oversight.

• 8.1SeminarAI policy and lawRead the rules that bind your deployment.3 hCompliance memo
• 8.2LabAgent securityDefend against injection and data leaks.4 hThreat model
• 8.3SeminarLabour and economicsModel who gains and who loses.3 hImpact note
• 8.4Field workPublic-sector deploymentShip under oversight.8 hDeployment plan
• 8.5Field workCapstoneShip to real users and defend it in an oral review.40 hLive product

S9 · 6 lessons · 49 h

Creative Studio

Make film, music and games with AI as your crew, while you keep the taste, the vision and the credit.

• 9.1StudioDirecting film with AITake a story from script to shot list to finished short, and direct every choice.8 hShort film with a process log
• 9.2StudioComposing music with AIWrite, arrange and produce a track where you decide what stays.6 hTrack with stems and notes
• 9.3LabDesigning games with AIPrototype a playable loop with AI as your assistant designer and tireless playtester.8 hPlayable prototype
• 9.4SeminarWorldbuilding and story systemsBuild a consistent world, its characters and a branching story, with a continuity check.4 hWorld bible
• 9.5SeminarRights, provenance and creditKnow who owns what, whose voice or likeness you may use, and how to credit it.3 hRights and credits sheet
• 9.6Field workRelease a workShip something to a real audience and learn from how they respond.20 hPublic release with audience notes

Hours are estimates, not promises. The artefact is what counts.