Khaled G.

Most automation freelancers build workflows. I build automation infrastructure—systems with rollout plans, error recovery, audit trails, and documentation your internal team can own after day one. That distinction matters when you're spending a serious budget on a project that needs to work in production, not just in a demo. Who I am: Telecommunications engineer by training (SDR, FPGA, signal systems). Founder of Elevorix Solutions—an AI/ML automation consultancy and training firm operating across the MENA region. Our n8n curriculum is n8n-approved, which means the architecture and methodology I bring to client projects are the same framework we teach, certify, and have validated at the platform level. I didn't start on Upwork—I built Elevorix's own internal automation stack first. That means everything I offer clients has already been stress-tested internally before it goes near a production environment. What I've actually built: Field Engineering O&M Automation — 12-node n8n workflow covering asset tracking, maintenance scheduling, and escalation routing. Delivered with a full 8-phase rollout plan: environment setup → staging → pilot → team handoff → monitoring cadence. Not a workflow file — a deployment program. Gmail Intelligence System — 24-node n8n pipeline integrating Gemini 2.0 Flash for email classification, priority tagging, and routing. Outputs a structured daily executive summary in HTML, logged to Google Sheets for audit. Processes hundreds of emails per day with zero manual intervention. WhatsApp Business API Engagement Platform — Full broadcast and auto-reply system built on n8n with opt-out compliance logic, contact segmentation, and Supabase as the backend datastore. Handles real message volume at scale. What I build for clients: → Multi-LLM AI agent orchestration inside n8n—route tasks intelligently between Claude, Gemini, and GPT based on cost, latency, and capability → CRM automation — lead scoring, routing, follow-up sequences, and pipeline triggers without touching your CRM's native limitations → Customer engagement automation — WhatsApp, email, and webhook-based systems that handle real conversation volume → Self-hosted n8n on Docker — proper deployment with health checks, alerting, and recovery workflows built in from the start → Full documentation and handoff — every project includes SOPs, annotated workflow maps, and a runbook. Your team inherits something they can actually maintain. Why this matters for your proposal: If you've been burned by automations that break when something unexpected happens or freelancers who deliver a workflow file and disappear—this engagement model is designed around the opposite outcome. I scope carefully, build with failure states in mind, and don't consider a project done until the person running it after me understands it. If you need automation that survives contact with a real enterprise workflow — and the documentation to prove it — let's talk about your use case.