Back to Jobs
Shopify

AI Systems Engineer

Confidential employer

Posted 10/23/2026Rate visible after sign-upSource: Curated from source
AI Tools
Shopify
Notion
Research
About the Role

Hey future team member, glad you're here.

We run several e-commerce brands and build our own AI systems every day: an autonomous video editor that cuts ads from voiceover and b-roll, a static ad engine that uses every data point we have, and AI agents that research competitors and products. We're hiring an extremely skilled engineer to help us build them.

How it works: we explain in detail how a system should work, and you build it. You ask sharp questions, push back when something doesn't add up, and deliver tested code that actually runs.

Who we're looking for:

  • 3+ years building real software in Python.
  • Hands-on experience with LLM APIs (Claude, OpenAI, Gemini): agents, tool calling, structured output.
  • Heavy AI coding tool user. Claude Code, Cursor, Codex. We expect you to use AI to work fast.
  • Solid with APIs, git, the command line, and debugging on your own.
  • You test your own work before you call it done.
  • Fluent written English.
  • Big plus: MCP, ffmpeg/video processing, Shopify or Meta Ads API.

What you'll build:

  • AI agent pipelines, including agents that check other agents' work.
  • Video and image pipelines with Python, ffmpeg, and AI models.
  • Integrations with Shopify, Meta, Notion, Google Drive, and Slack.
  • Automations with monitoring that alert us when something breaks.

We are good at AI, we need a master Engineer.

What we offer:

  • A long-term, full-time role working directly with the founder.
  • Hard, interesting problems at the frontier of AI and e-commerce.
  • Paid AI tool subscriptions.
  • Pay that grows with the quality and value of your work.

How to apply:

Start your message with "Agent builder". Applications without this will be ignored.

Include:

1. A link to ONE thing you've built with LLMs (GitHub, demo, or screen recording) and a short note on what you did.
2. A 2-minute Loom introducing yourself and walking through that project.
3. In 3–5 sentences: an AI agent is wrong 10% of the time, but the output looks correct so nobody notices. How would you build the system so those errors get caught automatically?

Make sure all links work without requesting access.

Looking forward to meeting you!

Ludvig Thunberg
Vetpets | Kiddio | Hermony

Track this external role

Sign in to save this listing and apply through the original source.

Sign in to SaveReport JobOpen Original Listing