What I'm focused on right now.
Scaling the Real User Monitoring platform with an event-driven pre-aggregation pipeline across 34 Turso/SQLite tables, Bun/Elysia backend on Cloudflare Workers, and a Next.js dashboard. Focused on keeping per-request costs near zero while handling high-throughput telemetry ingestion.
Architected a full-stack live classroom platform with LiveKit replacing Zoom — handling 10+ concurrent classes and 200+ simultaneous students. Recently cut dashboard load time by 94% (52s → ~3s) through server-driven MongoDB queries, index optimization across 100+ models, and eliminating BSON deserialization bottlenecks.
Published a guide on building a self-hosted reverse port-forwarding tool in Go (Go Tunnel), and a practical approach to creating an MDX-powered blog with GitHub and Bun. Maintaining Go Tunnel and building privacy-first developer tools.
Deep-diving into event-driven architectures, microservices patterns, and large-scale database optimization. Recent work includes MongoDB aggregation pipeline tuning (25% load reduction), compound indexing on 16 models, and cutting read-heavy endpoint latency by 40-60% across a production platform.
Last updated: July 10, 2026. prabhatlabs.dev © 2026