Repository for designing, iterating, and delivering MongoDB education labs for human learners. Learners work through progressively harder labs that teach MongoDB concepts by doing — starting from SQL thinking and migrating to idiomatic MongoDB patterns.
This repository is the counterpart to agent-training. Where agent-training optimizes for external AI agents with explicit scaffolding and zero implicit context, this repository optimizes for human learners with storytelling, motivation scaffolds, and struggle recovery cues.
| Directory | Purpose |
|---|---|
agents/ |
AI agents that automate the lab design and build workflow |
standards/ |
Instructional design principles and rubrics for human-centered MongoDB education |
skills/ |
Reusable domain knowledge modules (teaching strategies, patterns, etc.) |
labs/ |
Lab outlines, technical specs, and evaluation reports |
lab-test-env/ |
Self-contained lab environments (Docker, Node.js, check scripts) |
docs/ |
Research, design thinking, and learner feedback |
| Lab | Status | Difficulty |
|---|---|---|
| Builder Badge | Outline | Intermediate |
| (Your next lab here) | — | — |
Labs are created using a 5-agent pipeline (see agents/AGENTS.md):
1. /outline-builder — Design high-level outline
2. /converter — Convert outline → technical spec
3. /spec-reviewer — Validate spec (gate: ≥8/10)
4. /lab-builder — Build lab environment
5. /qa-review — QA review for learner experience
Read standards/README.md for pedagogical principles and the lab design template.
Each lab has a README.md in its lab-test-env/{lab-name}/ folder with setup and execution instructions.
See docs/ for guidance on running these labs in classroom, bootcamp, or self-paced contexts.
| Aspect | agent-training | human-training |
|---|---|---|
| Audience | External AI agents | Human learners |
| Scaffolding | Explicit, exhaustive | Strategic (some struggle is productive) |
| Instructions | No implicit context, everything stated | Assumes common sense, domain knowledge |
| Feedback | Mechanical pass/fail checks | Formative feedback with explanations |
| Pacing | Agent-speed (fast completion) | Human-speed (time for reflection) |
| Motivation | Performance metrics, mastery | Curiosity, real-world application, community |
| Error Recovery | Restart; try again | Guided debugging, partial credit |
We welcome contributions:
- New labs — design and submit via pull request (see standards/ for the template)
- Feedback — run a lab and tell us where you got stuck or what was unclear
- Bug reports — if a lab environment breaks or a check fails unexpectedly
- Ideas — open an issue to discuss MongoDB concepts we should teach
All labs follow the Instructional Design Rulebook. See that document for core learning science principles and how we apply them.
- Lab design questions — see
standards/andskills/ - Environment setup issues — see each lab's README under
lab-test-env/{lab-name}/ - General feedback — open an issue
Repository: danielcurran/human-training
Companion Repo: danielcurran/agent-training
License: MIT (for code and environment); instructional content TBD