In September 2026 I built a System Health Dashboard, solo. It is live: a self-service status page for customer support, the CTO and on-call engineers.
Context
Customer support, the CTO and on-call engineers needed one place to see whether the business-critical parts of the platform were healthy, without having to ask the tech team.
My role
I built it alone.
Approach & trade-offs
- A read-only agent runs the checks every few minutes, with no application code changes.
- Business-level checks across access, trading, payments, notifications and KYC, rather than raw infrastructure metrics.
- Plain-English guidance on what each status means and what to do, plus history.
Results
- ~80%fewer support questions to tech (estimate)
- Day onecaught live issues
- 0app code changes
Support questions to the tech team dropped about 80% (estimate). On day one it caught provisioning and payment-gateway issues.
More case studies
- Real-Time API Alerting on Cloudflare: Catching Issues Before Customers Do
- Affiliate Portal: Architecture, Deployment Security and CI/CD for an In-House Platform
Work with me: SRE and reliability reviews · Hire or work with me · My background