n8n automation & development
Production-grade workflows from a verified n8n creator — custom nodes, AI agents, and rescues for automations that outgrew Zapier, Make, or their first builder.
n8n is where automation starts. Cloudflare is where it can't fail. I build the durable-execution layer — Workers, Workflows, queues, and AI-agent infrastructure — that runs at scale without falling over. It's the same engineering behind a 25-worker production platform I own and operate.
CLOUDFLARE WORKERS
IN PRODUCTION
LIVE PRODUCTS USING
AUTOMATIONS I BUILT
NODES IN MY LARGEST
n8n PRODUCTION WORKFLOW
HOURS SAVED MONTHLY
BY A TYPICAL BUILD
HarperFlow runs your company blog end to end. It learns your site, finds the topics worth owning, then researches 8–12 sources per article, writes, fact-checks, and publishes on a schedule — straight into your Webflow CMS.
Production-grade workflows from a verified n8n creator — custom nodes, AI agents, and rescues for automations that outgrew Zapier, Make, or their first builder.
Workers, Workflows, and durable execution that run at scale without falling over — the same infrastructure behind a 25-worker production platform.
Autonomous agents that research, decide, and execute with tools, memory, and guardrails — built to run unattended, with human review only on exceptions.
Enrollment, content, student Q&A, and certificates automated end to end — powered by proprietary LMS code productized through CourseKit.
Keyword-to-published content engines that research, write, fact-check, and publish on a schedule — at scale, without making slop.
When visual tools hit their limit: custom code, self-hosted infrastructure, and APIs I own and operate — not black boxes I rent.
n8n, Cloudflare Workers, Claude, and the rest of the stack — orchestrated into one system that ships to production, not a demo.
Fixed-price, escrow-protected, and scoped up front. No hourly meter, no surprise invoices — you know the number before we start.
Production systems on Workers, Workflows, and Durable Objects — durable execution, queues, cron, and AI-agent infrastructure that runs unattended, at scale.
Workflows that survive restarts, retries, and failures without losing state — so a long-running or multi-step AI pipeline resumes exactly where it stopped instead of starting over or dropping data.
Yes — a common project. n8n is ideal for building fast; when a workflow needs production durability, I port the critical path to Workers and Workflows with an HMAC handoff, keeping n8n where visual speed still wins.
Production infrastructure starts at $15k+. Failure here is expensive and the skill is scarce — you're buying reliability and autonomy that compound, not billable hours. Most start with a $1,500 diagnostic audit, credited toward the build.
Yes — agent systems with tools, memory, and guardrails on Workers AI and the durable-execution primitives, built to run reliably in production, not just a notebook.
Code-first durable execution is a narrow skill. I run a 25-worker platform of my own on exactly this stack — so you get someone who operates it daily, not a team learning it on your budget.
Whatever the job needs — Workers and Workflows at the core, plus n8n for orchestration, real code, and the integrations around it. One connected system that ships to production.
US-registered (Better Brain Lab LLC, Sheridan, WY); fully remote, clients worldwide. Chubb-insured and NDA-ready.
Got infrastructure that simply can't fail?
Book an infrastructure call
Impossible workflows, built end to end.
n8n where speed wins. Cloudflare where production counts.
