Darryl F.

I'm a Production Systems Engineer who builds AI automation. I've spent years managing the kind of infrastructure most people don't think about until something goes down: servers, firewalls, production pipelines, enterprise networks. These days I bridge that background with AI automation. I build n8n workflows connected to Claude and OpenAI, deploy VAPI voice agents that handle live inbound calls, and wire everything into CRMs like GoHighLevel so leads get qualified, appointments get booked, and records stay in sync. Because I come from infrastructure, I don't just build the workflow. I secure it, host it, and monitor it. That end-to-end ownership is the difference between a demo that works once and a system that runs reliably in production. Here's what I regularly deploy for clients: ────────────────────────────── AI AUTOMATION & WORKFLOW DEVELOPMENT ────────────────────────────── • N8N pipelines connecting CRMs, webhooks, Google Workspace, Slack, and REST APIs with proper error handling • AI agent workflows using Claude and OpenAI for document processing, call scoring, lead qualification, and multi-step decision logic • Self-hosted n8n deployment, database optimization, migration, and disaster recovery ────────────────────────────── VOICE AI & CRM INTEGRATION ────────────────────────────── • End-to-end VAPI deployment covering custom LLM prompting, call flow logic, and inbound call handling • GoHighLevel integration with VAPI for CRM contact updates from call outcomes, appointment booking, and calendar sync triggered by live call events • Fully autonomous pipelines where the voice agent qualifies the lead, books the appointment, and updates the CRM with no human intervention required ────────────────────────────── SECURE SELF-HOSTED INFRASTRUCTURE ────────────────────────────── • Proxmox VE with isolated VMs for multi-tenant or dedicated client environments where each workload stays completely separated • Docker stacks on bare-metal servers (Hetzner/OVH) with Caddy reverse proxy, automated SSL, and NetBird/WireGuard mesh VPNs • Virtualized pfSense firewalls with Suricata IPS and pfBlockerNG for perimeter-level threat blocking ────────────────────────────── OBSERVABILITY & DISASTER RECOVERY ────────────────────────────── • Prometheus + Grafana monitoring so infrastructure problems get caught before your clients notice them • Proxmox Backup Server with encryption and deduplication for disaster recovery that actually holds up when you need it ────────────────────────────── 100% Job Success Score. Every engagement ends with clean documentation so you own the system completely and aren't locked into a single developer. If you're building a resilient AI automation system or need someone to fix a broken production pipeline, feel free to reach out.