05/06/2026
Paper-trading FXPulse on a Pepperstone demo account, building in public. Been diving into how we actually connect Claude to our trading stack—TradingView webhooks, API calls, the whole thing. Here's what I realized: most 'AI trading assistants' are just fancy automation dressed up in marketing. The real question is whether it survives contact with live markets.
Everyone pitches you the shiny integration. Nobody talks about the gap between "automated" and "automated that actually works when real money is on the line." We're building this in public specifically because that gap is where most things die. Paper-trading first. No promises until we've earned them. Just trying to be honest about what works and what doesn't.
Curious—have you tried building your own trading automation? What surprised you most about the difference between backtest and live?