AI Automation Specialist | Airtable · n8n · Make · GoHighLevel

Most Airtable setups i see are glorified spreadsheets. I design bases around how the business actually works. Most recent build: real estate acquisitions CRM for a US client. Four linked tables. Four role-specific interfaces. Webhook receiver handling DNC propagation, dedup logic, and unknown caller branching automatically. Lead scoring formula across five criteria. Paid, delivered, all test scenarios passing.

I build automations where AI does a specific job inside the workflow. Not a chatbot add-on. The decision logic. Lead scoring pipeline that qualifies and routes inbound leads in under 60 seconds. Voice agent that handles clinic calls end-to-end without a receptionist. Production Airtable CRM with a webhook receiver handling four distinct scenarios automatically. I diagnose the process first. If the foundation is broken, AI makes it break faster. I fix the foundation before i build on it.

REST API is how most of my builds talk to each other. My primary tool is n8n — i use it for complex multi-step workflows that need conditional routing, error handling, and webhook-based triggers. Builds: lead scoring pipeline, voice agent backend, multi-directory lead scraper with Apollo enrichment. Every workflow has error branches. Nothing fails silently. Also comfortable building direct API calls in Make, Zapier HTTP modules, and Google Apps Script where the platform doesnt have a native connector.

I use GHL as the CRM layer in multi-tool automation stacks. Contact creation, pipeline routing, tiered tagging, workflow automation, outbound sequence triggers. Built it into a lead scoring pipeline where inbound leads route by tier within 60 seconds. Hot leads get a Slack alert and a 30-minute follow-up reminder if no action. Cold leads go into drip sequences. No manual sorting.

I use Zapier for builds that dont need a full n8n workflow. Multi-source lead enrichment pipeline: Apollo.io pulls company data on form submission, Paths sort by priority, Slack fires an AI-generated two-sentence brief to the rep. 29-step client onboarding CRM triggered by Asana status changes, with automatic Drive folder creation, welcome emails with PDFs, and service-specific follow-up sequences. Filters, lookup steps, multi-step Zaps with conditional logic. Not just A-to-B connections.

Make Intermediate certified. Builds: Gmail attachment classifier (Gemini API classifies and renames files on arrival), Xero-to-Asana transaction sync (exports account data, formats as CSV, creates structured tasks), Gmail invoice watcher logging to Google Sheets. Comfortable with routers, iterators, error handling, and connecting APIs that dont have native Make modules.