Co-owner & CTO · Vibecode Ltd
/ˌtɛk.nɪ.kəl ˈliːd/ · noun
1.The one who owns the whole system — architecture, backend, and the infrastructure it runs on — and keeps it alive in production.
Co-owner & CTO at Vibecode Ltd. I design, ship, and operate production platforms end to end — async APIs, replicated databases, and the infrastructure beneath them. I make the architecture calls, then I'm the one on the pager.
§ Systems
Production systems I own end to end — and the platform engineering behind them. Each note is about the architecture and the calls behind it: replication, connection pooling, rate-limiting, observability, and staying up.
A B2B storefront and API serving a live business — architected, built, and operated end to end.
FastAPI async backend on a replicated PostgreSQL 16 core with streaming hot-standby; PgBouncer transaction pooling to survive connection storms; Redis for caching and distributed rate-limiting; load-balanced behind nginx and exposed through a Cloudflare tunnel with zero open ports. An LLM translation layer runs a local model with an OpenAI fallback.
An end-to-end production delivery pipeline — infrastructure-as-code all the way to a self-healing Kubernetes cluster on AWS.
Terraform provisions the AWS foundation (VPC, EKS, Multi-AZ RDS); a containerised Node + PostgreSQL service ships through GitHub Actions and Jenkins; Helm charts (dev / staging / prod) deploy via ArgoCD GitOps with auto-sync and self-healing; an HPA autoscales 2–10 pods, with Prometheus + Grafana for observability and RBAC / secrets / non-root containers for security.
A web control plane for the whole fleet — dashboard, in-browser terminal, and an AI agent layer that operates the infrastructure.
Next.js 15 (App Router, standalone) with a live xterm terminal over websockets and Mastra agents that automate operations. Every service streams metrics and logs to Prometheus, Loki, and Grafana for full observability.
The private cloud everything runs on — Proxmox infrastructure I architect and operate myself.
VMs and LXC containers, a dedicated streaming-standby database node, Docker Compose orchestration, and Cloudflare tunnels fronting every public service. Designed for production reliability and zero exposed ports on a single-operator budget.
§ Stack
The languages, data, and infrastructure I build, ship, and run production systems with.
§ Contact
Available to lead engineering — fractional CTO, technical lead, or backend architecture.
If you need someone to own the architecture and actually run what gets built, let's talk. Email is fastest.