Add AI capabilities to a MongoDB app including LLM summarization, structured generation, RAG pipeline with Atlas Vector Search, Voyage AI embeddings, and usage tracking with cost estimation

2026-03-028

mongodb-atlas-provisioning

mrlynn/claude-skills

Self-service Atlas cluster provisioning with HTTP Digest auth, Admin API v2 client, 9-step orchestration with rollback, status polling, and DevRel attribution tracking

2026-03-028

name	developer-day-curriculum
description	Generate workshop agendas and hands-on curriculum for customer developer days, technical training sessions, and field engagements
license	MIT
metadata	{"version":"1.0.0","author":"Michael Lynn [mlynn.org](https://mlynn.org)","category":"education","domain":"technical-training","updated":"2026-03-01T00:00:00.000Z","python-tools":"audience_analyzer.py, agenda_generator.py, exercise_scaffolder.py","tech-stack":"python, json, markdown"}

developer-day-curriculum

Trigger

Use this skill when planning customer developer days, field workshops, technical training sessions, or conference tutorials.

Trigger phrases:

"Plan a developer day for [customer]"
"Create workshop agenda"
"Design hands-on training"
"Build curriculum for [topic]"
"Generate developer day schedule"

Overview

Developer days require careful planning: the right topics, appropriate depth, hands-on exercises that work, and perfect pacing. This skill analyzes customer needs, generates time-blocked agendas, creates scaffolded exercises, and produces follow-up materials.

Not for: Self-paced courses or recorded content. This is for live, interactive workshops with hands-on components.

How to Use

Quick Start

Analyze audience:

python scripts/audience_analyzer.py customer-profile.json --output analysis.json

Generate agenda:

python scripts/agenda_generator.py analysis.json --duration 6 --output agenda.md

Create exercises:

python scripts/exercise_scaffolder.py agenda.md --output exercises/

Python Tools

scripts/audience_analyzer.py — Analyze customer tech stack, experience level, goals
scripts/agenda_generator.py — Generate time-blocked agenda with topics and breaks
scripts/exercise_scaffolder.py — Create hands-on exercise templates and solution code

Reference Docs

references/workshop-design-patterns.md — Proven workshop structures and pacing
references/customer-engagement-best-practices.md — Field engagement patterns from 50+ developer days

Templates & Assets

assets/agenda-template-halfday.md — 3-4 hour workshop template
assets/agenda-template-fullday.md — 6-8 hour workshop template
assets/exercise-template.md — Hands-on exercise structure
assets/feedback-survey.md — Post-workshop feedback template

Architecture Decisions

Why Audience Analysis First

Different audiences need different approaches:

Audience Type	Approach	Example Topics
Beginners	Guided, step-by-step	"Intro to MongoDB", "Your First Query"
Intermediate	Problem-solving focused	"Schema Design Workshop", "Aggregation Deep Dive"
Advanced	Architecture & optimization	"Performance Tuning", "Sharding Strategies"
Mixed	Layered (core + advanced breakouts)	Morning: Foundations, Afternoon: Tracks

Tool: audience_analyzer.py categorizes audience and recommends depth.

Time-Blocking Strategy

Attention spans are finite. Research shows:

10-15 min: Maximum lecture without interaction
45-60 min: Maximum before needing a break
90 min: Maximum hands-on exercise session

Agenda structure:

09:00-09:15  Welcome & Introductions (15 min)
09:15-09:45  Topic 1: Lecture (30 min)
09:45-10:30  Exercise 1: Hands-on (45 min)
10:30-10:45  Break (15 min)
10:45-11:15  Topic 2: Lecture (30 min)
...

Pattern: Lecture → Exercise → Break (repeat)

Hands-On Exercise Philosophy

Progressive Disclosure:

Starter code provided (80% complete)
Clear TODOs with hints
Solution available but not shown upfront
Extension challenges for fast finishers

Anti-pattern: Blank slate ("Build X from scratch"). Most participants fail.

Good pattern: Guided scaffolding with specific gaps to fill.

Pacing Rules

Duration	Topics	Exercises	Breaks
Half-day (3-4h)	2-3 topics	2 exercises	1 break
Full-day (6-8h)	4-6 topics	4-5 exercises	3+ breaks
Multi-day	8-12 topics	8-10 exercises	Lunch + breaks each day

Rule of thumb: 1 topic per hour (30 min lecture, 30 min exercise/discussion).

Follow-Up Materials

Participants forget 70% within 24 hours without reinforcement. Provide:

Slide deck (PDF, not editable)
Exercise repo (GitHub with solutions)
Resource links (docs, tutorials, videos)
Next steps (suggested learning path)
Contact info (Slack, email, office hours)

Generated Output Structure

Agenda (Markdown)

# MongoDB Developer Day - [Customer Name]
**Date:** [Date] | **Duration:** 6 hours | **Level:** Intermediate

## Goals
- [Learning objective 1]
- [Learning objective 2]

## Prerequisites
- Basic MongoDB knowledge
- Laptop with [requirements]

## Schedule

### Morning Session (9:00 AM - 12:00 PM)

#### 9:00-9:15 - Welcome & Introductions (15 min)
- Icebreaker
- Agenda overview
- Wi-Fi and logistics

#### 9:15-9:45 - Topic 1: [Title] (30 min)
**Learning objective:** [What participants will learn]
**Key concepts:** [Bullet list]

#### 9:45-10:30 - Exercise 1: [Title] (45 min)
**Goal:** [What participants will build]
**Starter code:** `exercises/01-topic-name/`
**Success criteria:** [How to know you're done]

...

## Follow-Up Resources
- [Links]

Exercise Structure

exercises/
├── 01-topic-name/
│   ├── README.md          # Instructions
│   ├── starter/           # Starter code (80% complete)
│   │   └── app.js
│   ├── solution/          # Complete solution
│   │   └── app.js
│   └── extension/         # Bonus challenges
│       └── challenges.md

Python Tool Details

1. Audience Analyzer

Purpose: Analyze customer profile and recommend topics/depth.

Input: Customer profile JSON

{
  "company": "Acme Corp",
  "industry": "FinTech",
  "audience_size": 25,
  "experience_level": "mixed",
  "current_stack": ["PostgreSQL", "Redis", "Python"],
  "mongodb_experience": "beginner",
  "goals": [
    "Migrate from PostgreSQL to MongoDB",
    "Learn schema design patterns"
  ],
  "constraints": [
    "6 hour duration",
    "Must include hands-on exercises"
  ]
}

Output: Analysis JSON

{
  "recommended_level": "intermediate",
  "suggested_topics": [
    "MongoDB fundamentals (quick refresher)",
    "Relational to document migration patterns",
    "Schema design workshop",
    "Aggregation pipeline deep dive"
  ],
  "pacing": {
    "lecture_ratio": 0.4,
    "exercise_ratio": 0.5,
    "break_ratio": 0.1
  },
  "focus_areas": [
    "Emphasize schema design (coming from relational)",
    "Include PostgreSQL → MongoDB migration example",
    "Address transaction concerns (FinTech requirement)"
  ]
}

2. Agenda Generator

Purpose: Generate time-blocked agenda from analysis.

Usage:

python scripts/agenda_generator.py analysis.json \
  --duration 6 \
  --start-time "9:00 AM" \
  --output agenda.md

Options:

--duration - Hours (default: 6)
--start-time - Start time (default: 9:00 AM)
--include-lunch - Add lunch break for full-day
--format - Output format (markdown, html, pdf)

Output: Markdown agenda with:

Time-blocked schedule
Topic descriptions
Exercise placeholders
Break timing
Prerequisites and setup instructions

3. Exercise Scaffolder

Purpose: Generate hands-on exercise templates from agenda.

Usage:

python scripts/exercise_scaffolder.py agenda.md --output exercises/

Generates:

Exercise directory structure
README with instructions
Starter code templates
Solution placeholders
Extension challenges

Example output:

exercises/
├── 01-schema-design/
│   ├── README.md
│   ├── starter/
│   │   ├── schema.js        # 80% complete with TODOs
│   │   └── sample-data.json
│   ├── solution/
│   │   └── schema.js        # Complete solution
│   └── extension/
│       └── challenges.md    # Bonus challenges

Workflow Example

Scenario: Plan a 6-hour MongoDB developer day for FinTech customer

Step 1: Create customer profile

{
  "company": "FinTech Corp",
  "audience_size": 20,
  "experience_level": "intermediate",
  "current_stack": ["PostgreSQL", "Python", "AWS"],
  "mongodb_experience": "beginner",
  "goals": ["Migration from PostgreSQL", "Schema design"],
  "constraints": ["6 hours", "Must include transactions"]
}

Step 2: Analyze audience

python scripts/audience_analyzer.py fintech-profile.json --output analysis.json

Output: Recommends intermediate level, focus on migration patterns and transactions

Step 3: Generate agenda

python scripts/agenda_generator.py analysis.json \
  --duration 6 \
  --start-time "9:00 AM" \
  --include-lunch \
  --output fintech-agenda.md

Output: 6-hour agenda with 4 topics, 4 exercises, lunch, 2 breaks

Step 4: Create exercises

python scripts/exercise_scaffolder.py fintech-agenda.md --output exercises/

Output: 4 exercise directories with starter code, solutions, extensions

Step 5: Customize

Fill in starter code with FinTech-specific examples
Add real PostgreSQL → MongoDB migration example
Customize transaction exercise for financial use case

Step 6: Deliver

Present agenda
Walk through exercises
Collect feedback
Share follow-up resources

Common Patterns

Pattern 1: The "Hook" Opening

Start with a compelling demo, not slides.

Bad:

"Welcome. Today we'll learn about MongoDB. Here are 30 slides..."

Good:

"Let me show you something cool." [Live demo: Build a real-time dashboard in 10 minutes] "That's what you'll build today. Let's get started."

Pattern 2: "You Try" Checkpoints

Every 15 minutes, pause for a quick "you try" moment.

Example:

"I just showed you how to create an index. Open your terminal and try it with your dataset. 2 minutes. Go!"

Prevents passive watching.

Pattern 3: Progressive Complexity

Start simple, layer on complexity.

Exercise 1: Basic CRUD operations Exercise 2: Add indexes for performance Exercise 3: Add aggregation pipeline Exercise 4: Add transactions and error handling

Each builds on the previous.

Pattern 4: Real-World Context

Use customer's industry for examples.

Generic (boring):

"Here's a user schema..."

Industry-specific (engaging):

"Here's how to model a financial transaction with MongoDB..."

Pattern 5: Fast Finisher Extensions

Always have bonus challenges ready.

## Exercise 1: Basic Schema Design

**Success Criteria:**
- [x] Created schema
- [x] Inserted sample data
- [x] Queried data

**Done early? Try these extensions:**
1. Add validation rules
2. Create compound indexes
3. Add data migration script

Keeps advanced participants engaged.

Timing Anti-Patterns

❌ Too Much Content

Cramming 12 topics into 6 hours. Result: Rushed, no hands-on time.

Fix: 4-6 topics maximum for 6 hours.

❌ No Breaks

Running 3 hours straight without a break.

Fix: Break every 60-90 minutes minimum.

❌ Long Lectures

45-minute lecture with no interaction.

Fix: 15-minute chunks with "you try" checkpoints.

❌ Blank Slate Exercises

"Build a complete app from scratch in 45 minutes."

Fix: Provide 80% starter code with specific TODOs.

❌ Skipping Setup Time

Jumping straight into exercises without verifying everyone's environment works.

Fix: 15-minute setup verification at start.

Quality Checklist

Before delivering:

When to Use Different Formats

Format	Best For	Example
Half-day (3-4h)	Conference tutorials, intro workshops	"Intro to Vector Search"
Full-day (6-8h)	Customer developer days, deep dives	"MongoDB Schema Design Workshop"
Multi-day (2-3 days)	Comprehensive training, bootcamps	"MongoDB Developer Bootcamp"
Workshop series	Ongoing enablement	Weekly 2-hour sessions over 4 weeks

Customization Tips

For Enterprise Customers

Use their industry terminology
Reference their specific use cases
Include compliance/security topics
Provide enterprise deployment examples

For Startups

Move faster, less hand-holding
Focus on practical "get it working" approach
Include scaling considerations
Share cost optimization tips

For Mixed Audiences

Morning: Core fundamentals (everyone)
Afternoon: Breakout tracks (beginner vs advanced)
Provide self-paced catch-up materials

Follow-Up Strategy

Immediately after (same day):

Share slide deck PDF
Send GitHub repo link with exercises/solutions
Collect feedback survey

Next day:

Send thank you email
Share additional resources
Offer office hours

One week later:

Check in on progress
Answer questions
Share related content

One month later:

Follow-up on original goals
Offer advanced workshop if interest

References

Workshop design patterns: references/workshop-design-patterns.md
Field engagement best practices: references/customer-engagement-best-practices.md
Adult learning theory: https://en.wikipedia.org/wiki/Andragogy
Hands-on exercise design: https://teaching.cornell.edu/resource/active-learning

Credits

Michael Lynn — mlynn.org · @mlynn · LinkedIn · GitHub

Next steps after generating curriculum:

Customize exercises with customer-specific examples
Test all hands-on exercises on fresh environment
Prepare backup demos (in case of connectivity issues)
Create feedback survey
Compile follow-up resources