client
Techs
Comprehensive IT Infrastructure & Operations Dashboard
The Problem
The enterprise IT team was juggling 4 separate tools (Grafana, Kibana, a custom alerting script, and Slack for escalation) — context-switching between them added 12+ minutes to mean-time-to-resolution during incidents.
The Solution
Built a single Vue.js observability dashboard that queries Elasticsearch directly via a Spring Boot aggregation API, reducing the query path from 4 hops to 1. Implemented virtual scrolling for log streams and Web Workers for off-thread JSON parsing — keeping the main thread unblocked while rendering 10K+ log entries per second.
My Role
Frontend & Platform Engineer
Timeline
7 months
Product
Architecture
Frontend
Vue.js & Tailwind
Backend
Java Spring Boot
Realtime
Polling & Webhooks
Infrastructure
On-Premises Kubernetes
Tech Stack
Impact
Implemented virtual scrolling for live telemetry feeds rendering 10,000+ log entries/second — offloading JSON parsing to Web Workers so the main thread stays unblocked, eliminating the UI freezes that were causing engineers to miss alert spikes during high-load events.
Consolidated 4 observability tools into one dashboard, reducing mean-time-to-resolution by 40% (from 18 minutes to 11 minutes) by eliminating context-switching between Grafana, Kibana, and the custom alerting script during incident response.