Level 7
10 min readLesson 39 of 43

The Go-Live Checklist

Everything to verify before flipping the switch

The Moment of Truth

Youve built the system. Youve tested it. Now comes the moment you flip the switch and let it trade real money. This checklist ensures nothing is forgotten.

Pre-Launch: Infrastructure

Exchange Accounts:

  • Production API keys created (not testnet)
  • IP whitelist configured
  • Withdrawal disabled on trading keys
  • 2FA enabled on exchange accounts
  • Sufficient funds deposited

Server:

  • Production server provisioned
  • System monitoring installed
  • Disk space adequate for logs
  • Automatic restart on crash configured
  • Firewall rules set

Network:

  • Static IP or dynamic DNS configured
  • Redundant internet if critical
  • VPN for secure access

Pre-Launch: System

Configuration:

  • All parameters set to production values
  • Risk limits configured and verified
  • Paper trading mode OFF
  • Logging level appropriate
  • Correct exchange endpoints (not testnet)

Database:

  • Production database provisioned
  • Backups scheduled
  • Connection pooling configured
  • Historical data loaded

Queues:

  • Redis running with persistence
  • Queue names set to production
  • Dead letter queues configured

Pre-Launch: Monitoring

Alerts Configured:

  • Error rate thresholds set
  • Position mismatch detection
  • System health checks
  • Notification channels verified (send test message)

Dashboards:

  • Real-time P&L visible
  • Position status visible
  • System health metrics visible
  • Log access working

On-Call:

  • Escalation contacts defined
  • Phone numbers verified
  • Backup contacts available

Pre-Launch: Operations

Documentation:

  • System architecture documented
  • Recovery procedures written
  • Contact information compiled
  • Common issues and solutions listed

Kill Switch:

  • Kill switch tested
  • Multiple people know how to use it
  • Works from mobile if needed

Runbooks:

  • Startup procedure documented
  • Shutdown procedure documented
  • Common issue resolution documented

Launch Day: Verification

Before enabling trading:

  1. Verify exchange connectivity (fetch balance)
  2. Verify data feeds working (check timestamps)
  3. Verify database writes working
  4. Verify notifications working
  5. Verify position tracking shows correct state
  6. Verify kill switch accessible

Launch Day: Staged Enablement

Dont go from zero to full immediately.

Hour 1: Enable one edge, minimum size Day 1: Monitor every trade manually Week 1: Gradually add edges, increase sizes Month 1: Full system, close monitoring

Launch Day: What to Watch

First Trade:

  • Did signal generate correctly?
  • Did order submit successfully?
  • Did fill occur at expected price?
  • Did position tracking update?
  • Did notification fire?

First Hour:

  • Error rate acceptable?
  • Latency within bounds?
  • No unexpected behavior?

Post-Launch: Immediate

Hour 1 Checklist:

  • Review all trades that occurred
  • Verify positions match exchange
  • Check for any errors in logs
  • Confirm monitoring working

Day 1 Checklist:

  • Comprehensive trade review
  • P&L calculation verification
  • System stability assessment
  • Adjust any problematic parameters

Post-Launch: Week 1

  • Daily trade reviews
  • Performance vs expectations
  • Bug identification and fixing
  • Parameter tuning if needed
  • Gradually increase confidence/size

Emergency Procedures

Know before you need them:

Exchange Unreachable:

  1. Check exchange status page
  2. Verify network connectivity
  3. If prolonged, consider manual intervention

Position Mismatch:

  1. Trust exchange state
  2. Reconcile local state
  3. Investigate cause

Unexpected Large Loss:

  1. Kill switch immediately
  2. Verify actual positions
  3. Investigate cause before resuming

Takeaway

Going live is a process, not an event. Methodical verification prevents expensive mistakes.

Resist the urge to skip steps. The one you skip is the one that bites you.

Next: what to watch in the critical first 30 days.