AI Automation Engineer (20h/week)

About the role

I’m hiring a reliable, proactive Automation Engineer to build and maintain automations and AI workflows for clients and internal products. You’ll work directly with me (no client calls). Your focus is delivery: build, test, improve, document.

We start with 20 hours per week (4 hours per day). If it’s a great fit, this can grow.

What you’ll do

- Build automation workflows end to end (design, build, test, ship)
- Integrate tools and systems (CRMs, Google Workspace, Notion, databases, web apps, Slack, WhatsApp etc.)
- Build AI-powered workflows, including RAG systems (retrieval + generation) and agent-style automations
- Work with databases like Supabase (or similar) to store, query, and structure data for automations
- Write code when needed (scripts, API integrations, transformations, custom logic)
- Maintain and improve existing automations (monitoring, fixing issues, making workflows more stable)
- Document every build clearly so I can understand, manage, and reuse it

Tech you’ll likely use

This is my current stack, but I’m open to better approaches if you can explain why and deliver faster or more robust solutions.

- Automation: n8n (common), other tools are welcome if you can justify them
- Databases: Supabase (preferred)/ pine cone, or similar modern DBs
- AI tooling: OpenAI, ChatGPT, Gemini, Copilot, claude
- Development: Cursor, python, scripting, APIs, webhooks
- Typical tools: Google Workspace, Notion, HubSpot and other CRMs

Must-haves

- Strong automation engineering skills, you can build complex workflows with or without low-code tools
- Solid software development fundamentals, you can code and think in systems
- Experience building RAG systems in real projects (not just theory)
- Comfortable working with databases (Supabase or similar), you can model data and query it
- Confident with API integrations and webhooks
- Excellent English communication, async-first, clear updates
- High reliability: fast response times, consistent availability, no disappearing

Nice-to-haves

- Experience with multi-step AI pipelines (classification, extraction, enrichment, summarization)
- Scraping and data cleaning experience
- Strong debugging mindset, you enjoy finding root causes and preventing repeats

What “great” looks like

- You take ownership: you spot risks and propose improvements before building
- You ship work that is stable, testable, and easy to maintain
- You communicate clearly and quickly, even when something breaks
- You think in reusable building blocks, not one-off hacks

Working style

- No client calls
- We do 2 short check-ins per week
- You work from a standard brief template I provide
- Updates via Slack or email

Dealbreakers

- Poor communication or slow responses
- Unreliable availability
- Sloppy work, no testing, no documentation

How to apply

Send:

- Your portfolio or 2 to 3 examples of workflows/projects you built (automation + code)
- A short Loom video (3 to 6 minutes) walking through one project:
What problem it solved
How you structured it
What could break, and how you handled that
Your weekly availability and when you can start