Skip to main content

Mike's Infrastructure Learning Path

Start Here

This is your 3-hour hands-on learning path for the infrastructure we deployed on 2026-02-03.

Time needed: 2-3 hours to learn, 2-3 minutes per use after that.

Goal: Make infrastructure usage automatic and habitual.

What We Built (30-Second Overview)​

You now have 6 production systems that are MANDATORY for all work:

  1. Decision Management - Document all architectural decisions as markdown files
  2. Sync Operations - Get data from Asana/Shortcut/Linear (no custom code)
  3. Rate Limiting - Automatic (prevents API errors)
  4. Memory & Research Notes - Store research findings (AI platform memory + notes files)
  5. Session Continuity - Continue work across machines via handoff files
  6. WPEngine Monitoring - 228 sites, automatic alerts

Why mandatory? Prevents wheel reinventing, enables cross-session learning, eliminates API errors.

Your Learning Path (3 Hours)​

Hour 1: Complete Manual Deployment​

Status: 71% deployed, needs 3 manual steps (65-75 min)

Why first? Can't use infrastructure until it's 100% deployed.

Step 1: Deploy Memory System (10-15 min)​

# Open the guide
open ~/GitHub/flypilot/docs/planning/MANUAL-DEPLOY-memory-system-2026-02-03.md

# Open Supabase SQL Editor
open "https://supabase.com/dashboard/project/wkydccpqyebmjzdwkjxe/sql/new"

# Copy SQL from guide → Paste in editor → Run (2 migrations)

# Test it worked
cd ~/GitHub/flypilot/docs/planning
./memory-system-quick-test.sh

Step 2: Fix 1Password Authentication (15-20 min)​

# Open the guide
open ~/GitHub/flypilot/docs/planning/1password-sa-setup.md

# Create service account: fp-tech-automation
# Grant Engineering vault access
# Update .env file

Step 3: Deploy WPEngine Alerts (40 min)​

# Open the guide
open ~/GitHub/flypilot/docs/planning/NEXT-STEPS-wpengine-ssl-2026-02-03.md

# Test SSL filtering
# Deploy to cron
# Monitor first run

Checkpoint: All systems operational ✅

Hour 2: Learn the Mandatory Workflow (Hands-On)​

Now you'll actually USE the infrastructure with a real task.

Pick a real task (not a test): Something on your actual to-do list.

Example: "Review today's infrastructure deployment"

The 5-Step Mandatory Workflow:​

# Step 1: SEARCH FIRST (don't reinvent wheels)
cd ~/GitHub/flypilot
grep -ril "infrastructure" docs/planning/decisions/
grep -ril "deployment" docs/planning/decisions/

# You'll see decision files like:
# docs/planning/decisions/2026-02-03-deploy-cron-to-mac-studio.md
# docs/planning/decisions/2026-02-03-store-credentials-in-1password.md

# Step 2: CHECK HANDOFFS (pick up ongoing work)
ls handoffs/

# Step 3: DO THE WORK
# ... actually do your task ...
# For this example: Review the deployment docs, test a command, etc.

# Step 4: RECORD YOUR DECISION (share knowledge)
cat > docs/planning/decisions/$(date +%Y-%m-%d)-adopt-automated-infrastructure-workflows.md << 'EOF'
# Decision: Adopt automated infrastructure workflows

**Date:** 2026-02-25
**Status:** Active
**Impact:** critical
**Category:** process
**Department:** fp

## Decision

All work sessions must follow the mandatory infrastructure workflow:
search decisions first, use sync profiles, document findings.

## Rationale

Prevents wheel reinventing, enables cross-orchestrator learning.

## Alternatives Considered

- **Manual ad-hoc approaches:** Leads to duplicated work and lost context
- **No enforcement:** Guidelines not followed without mandatory structure
EOF

# Step 5: DOCUMENT KEY FINDINGS (preserve research)
echo "- $(date +%Y-%m-%d): Infrastructure deployment completed 2026-02-03: systems operational, 21 agents executed — tags: infrastructure,deployment,milestone" >> docs/planning/research-notes.md

# Verify it worked
ls -lt docs/planning/decisions/ | head -5
grep "2026-02-03" docs/planning/research-notes.md

Practice this workflow 2-3 times with different real tasks until it feels natural.

Tasks to practice with:

  1. Review FP-Librarian's documentation work
  2. Plan next week's infrastructure training
  3. Review a client request

Time: 15-20 min per task → 45-60 min total

Checkpoint: You've used the workflow 3 times ✅

Hour 3: Explore the Infrastructure (Self-Guided)​

Now explore what's available.

1. Decision Management​

cd ~/GitHub/flypilot

# List all existing decisions
ls docs/planning/decisions/

# Search by keyword
grep -ril "database" docs/planning/decisions/
grep -ril "sync" docs/planning/decisions/
grep -ril "rate" docs/planning/decisions/

# Read a specific decision
cat docs/planning/decisions/2026-02-03-use-supabase-postgresql.md

Try it: Search for something you're working on. Did someone already make a decision about it?

2. Sync Operations​

cd ~/GitHub/flypilot
source .venv/bin/activate

# See all profiles
python -m sync.cli profiles

# Run a quick sync (dry-run first)
python -m sync.cli sync --profile quick --dry-run

# Run for real
python -m sync.cli sync --profile quick

# Department-specific
python -m sync.cli sync --profile omg  # OMG only
python -m sync.cli sync --profile lrnd   # LRND only

# Check the logs
tail -100 ~/logs/flypilot-sync.log

Try it: Run a quick sync and watch the rate limit statistics at the end.

3. Memory & Research Notes​

cd ~/GitHub/flypilot

# Add a finding to the research notes
echo "- $(date +%Y-%m-%d): Important fact I just learned — tags: category,topic" \
  >> docs/planning/research-notes.md

# Search existing findings
grep -i "keyword" docs/planning/research-notes.md

# Review all notes
cat docs/planning/research-notes.md

Try it: Document something you learned today. Search for it. Does it come back?

For Claude Code sessions: use /memory to manage project-level memory that persists across conversations.

4. Session Continuity​

cd ~/GitHub/flypilot

# Check for active handoffs
ls handoffs/

# Review a handoff if one exists
cat handoffs/session-Tech-Director-2026-02-24.md

# Create a handoff for your current work
cat > handoffs/session-$(date +%Y-%m-%d).md << 'EOF'
# Session Handoff

## Context
Exploring infrastructure systems during learning path.

## Completed
- Reviewed sync profiles
- Documented first research finding

## Next Steps
- Test WPEngine monitoring
- Record first architectural decision
EOF

Try it: Create a handoff file, end the session, start fresh and see how quickly you can pick up where you left off.

5. WPEngine Monitoring​

# Check what's being monitored
tail -100 ~/logs/flypilot-wpengine.log

# Check alerts
tail -100 ~/logs/flypilot-wpengine-alerts.log

# Query dashboard views (after Supabase deployment)
psql $SUPABASE_DB_URL -c "SELECT * FROM v_wpengine_health_summary LIMIT 10;"

Try it: Check the logs. Are 228 sites being monitored? Any alerts?

Checkpoint: You understand what each system does ✅

What Success Looks Like​

After 3 hours of learning:

  • ✅ All systems deployed (100%)
  • ✅ Workflow practiced 3+ times
  • ✅ Infrastructure explored and tested
  • ✅ Quick reference printed/bookmarked

After 1 week of daily use:

  • ✅ 10+ decisions recorded as files
  • ✅ Research notes being kept
  • ✅ 0 rate limit errors
  • ✅ 0 infrastructure bypasses
  • ✅ Workflow feels automatic

After 1 month:

  • ✅ 30+ decisions (knowledge base growing)
  • ✅ Research notes library is genuinely useful
  • ✅ All active orchestrators follow the workflow
  • ✅ Infrastructure is muscle memory

Common Questions​

Q: "This feels like overhead. Is it worth it?"

A: Workflow overhead is 2-3 min/session. Time saved: 30-60 min/session (not redoing work).

ROI: 10-20x return on time invested.

Q: "What if I forget a step?"

A: Every orchestrator role file has "Before Every Session" checklist. It's there as a reminder.

Q: "What if no one else is using it?"

A: You're the owner. When you use it, everyone else will follow.

Q: "Can I customize the workflow?"

A: Yes, but start with the standard workflow for 2 weeks first. Then optimize based on experience.

Q: "What if I make a mistake in a decision file?"

A: Edit the file and commit the update. Change the status to Superseded if a newer decision replaces it. Nothing is permanent — git tracks the history.

Next Steps After Learning​

  1. Day 1 Afternoon: Demonstrate the workflow to 1-2 key orchestrators
  2. Days 2-5: Train 10 orchestrators systematically
  3. Day 8: Complete first weekly review
  4. Week 2: Measure impact and refine

Getting Help​

Documentation:

Testing:

  • All sync commands have --help flag
  • All sync commands have --dry-run mode (safe to test)

Questions:

  • Search decisions first: grep -ril "your question" docs/planning/decisions/
  • Check research notes: grep -i "your question" docs/planning/research-notes.md

You're ready! Start with Hour 1 (deployment), then Hour 2 (practice), then Hour 3 (explore).

By dinner time today, you'll have used all infrastructure systems and the workflow will start feeling natural.