The Operating Challenge
Relying on third-party SaaS chat UIs incurs recurring subscription fees, routing bugs, and heavy JavaScript bundle payloads (>10 MB) that degrade mobile experience.
Personal Infrastructure & Lab · 2026 → Present01. Problem & Context 02. Architecture & Solution 03. Measurable Impact 04. Technology Stack
Cloud-Hosted OpenAI-Compatible LLM Gateway & Custom Chat UI
Cloud-hosted OpenAI-compatible LLM gateway and custom 15 KB static chat SPA on ARM64 cloud architecture, featuring Caddy reverse proxy, PocketBase forward_auth, and streaming SSE responses.
Technical Implementation
Architected a lean AI gateway stack: Caddy reverse proxy managing automatic Let's Encrypt TLS, PocketBase handling custom forward_auth verification hooks, and a 15 KB vanilla JavaScript SPA supporting SSE streaming, model switching, and document attachments.
Results & Verification
Achieves sub-50 ms API authorization and sub-second cold page loads on lean ARM64 cloud infrastructure.
Ask Bilal’s AI Clone