Data Architecture — 2023
Fish Count Platform
Client: ADF&G (Public Sector)
Problem
Real-time biological monitoring data on the Kenai River was siloed and difficult for the public to access during critical peak run cycles.
Approach
Designed a high-availability dashboard that polls state-level sonar data and renders it via a static-first React architecture. Reliability was the secondary objective—latency the primary.
Execution
Built using Vite and Tailwind CSS, deployed via Cloudflare's edge network. The system uses zero-egress R2 storage for asset delivery and aggressive ISR (Incremental Static Regeneration) logic to maintain sub-100ms response times.
Outcome
Successfully handled seasonal traffic spikes of 50k+ daily users without a single service interruption. Established as the authoritative source for Kenai River sonar metrics.
// System Telemetry
- FrameworkReact / Vite / Bun
- InfrastructureCloudflare Workers & R2
- Styling EngineTailwind CSS v3.4
- Data SourceADF&G Public API Polling
< 100ms
Response
100%
Uptime
50k/day
Load