Learn AI and compose a useful path
Use the twenty-lesson school as a dependency-aware route, then turn the relevant lessons into a portable instruction pack for your own agent.
TTC-101TTC-102TTC-103TTC-104TTC-105TTC-106TTC-107
Open manual →Public manuals · 5 stable guidesUse the method.
Use the method.
Inspect the history.
From first AI concepts to a portable private classroom: practical, versioned instructions with visible boundaries and no account required.
PUBLICATION CONTRACT20 of 20 lessons mapped.Human-reviewed versions.Private infrastructure excluded.HTML · Markdown · JSON
Five working manualsPractice over
Practice over
mystique.
Each manual maps to stable lesson IDs, includes reusable checks, and states what it cannot authorize.
Build context, rules, examples and visible memory
Separate temporary context, retrieved evidence, durable reviewed memory and model fine-tuning, then express reusable teaching as inspectable files.
TTC-108TTC-109TTC-110TTC-111TTC-112
Open manual →Version, test and review agent behavior
Treat changes to agent instructions and memory like changes to software: small revisions, visible diffs, reproducible tests and explicit acceptance.
TTC-113TTC-114TTC-115
Open manual →Bound permissions and consequential action
Give an agent only the information and tools required for the current purpose, and keep high-consequence actions behind explicit human control.
TTC-116TTC-117TTC-118
Open manual →Maintain freshness and run a portable classroom
Check trusted sources on a bounded schedule, keep editorial review separate from detection, and package the agent files so another compatible workspace can inspect them.
TTC-119TTC-120
Open manual →Authorship and source trailNamed work.
Named work.
Clear boundaries.
Authored and directed by Finn Andre Hotvedt. Developed with assistance from ChatGPT and Codex by OpenAI. Manual versions are published on this site as HTML, Markdown and JSON.
View the dedicated Teach manuals repository on GitLab ↗
This repository preserves the manuals and their real Git history. The complete Apache-2.0 application and Docker source is linked from the self-host page.
Complete subject mapAll twenty lessons,
All twenty lessons,
one visible route.
Use the school for the full lesson, sources and exercise. Use a manual when you want the operational workflow.
- TTC-101AI capabilities and limitsbeginner
- TTC-102Clear prompts and task briefsbeginner
- TTC-103Verification and source readingbeginner
- TTC-104Everyday privacy and data minimizationbeginner
- TTC-105Human decisions and approvalbeginner
- TTC-106AI for learning and studyingbeginner
- TTC-107Multimodal literacy: images, audio, and textbeginner
- TTC-108Teach one agent a bounded rolepractitioner
- TTC-109Knowledge and provenancepractitioner
- TTC-110Examples, counterexamples, and correctionspractitioner
- TTC-111Visible memory and forgettingpractitioner
- TTC-112Structured rules, YAML, and AGENTS.mdpractitioner
- TTC-113Version history and rollbackpractitioner
- TTC-114Creative collaborationpractitioner
- TTC-115Workplace task designpractitioner
- TTC-116Least privilege and prompt injectionadvanced
- TTC-117Evaluations and regression testsadvanced
- TTC-118Context, retrieval, visible memory, and model-weight fine-tuningadvanced
- TTC-119Responsible deployment and accountabilityadvanced
- TTC-120Portable and self-hosted AI workflowsadvanced