Make.com Engineer for AI-Assisted Event Data Pipeline

I’m building Scout, a personal passion project that helps people discover local events without searching dozens of websites. I’m a non-technical owner, so I need someone who can communicate clearly and explain recommendations in plain language. Scout has a curated library of 70+ public event sources across two categories. These sources fall into a small number of collection methods, so the goal is to build reusable collectors—not a separate automation for every website. Some sources will remain manual or be skipped. Sources include: Websites and public event calendars iCal/ICS feeds Eventbrite and similar platforms Email newsletters used for quality checking A working Make.com pipeline already collects events from an ICS feed, checks for duplicates, and writes them into Google Sheets. The project also has a documented architecture, data model, source library, and detailed blueprint for the next collector. First Paid Phase I need someone to: Review the existing documents and Make.com setup Confirm whether the proposed approach is sensible and recommend any necessary changes Build a reusable Make.com collector that fetches event pages and uses OpenAI or Claude to extract structured event information Test it on three different real websites, including a paginated site and a page containing both events and marketing content Add duplicate checking and practical error handling Leave brief, plain-language documentation that I can understand The initial review and planning should take approximately 5–10 hours. After that, we will agree on the scope and estimated hours for the build before continuing. Budget: $10–$25/hour, with an expected rate around $20/hour depending on experience. Ideal Experience Strong hands-on Make.com experience Extracting data from varied websites HTTP requests, APIs, webhooks, pagination, and structured JSON OpenAI or Claude extraction inside automated workflows Google Sheets automation Duplicate detection, error handling, and failure alerts Python, Playwright, or browser automation for JavaScript-heavy websites is helpful This is a real but modestly funded personal project, not a funded startup. I care more about a reliable, understandable solution than an unnecessarily complex one. Please begin your proposal with the word Scout and briefly answer: What is the most similar recurring Make.com or multi-source data pipeline you have personally built? Have you used OpenAI or Claude to extract structured information from inconsistent webpages? How did your system handle pagination, duplicates, website changes, and extraction failures? How would you handle websites whose event information does not appear in the raw HTML? Are you comfortable reviewing and improving an existing documented plan rather than starting from a blank page?