Any
1000-2500
TBD
Apr 21, 2026
Claude Code Power User & AI Automations Engineer — Full-Time, U.S. Company
We're a U.S.-based company running multiple brands across real estate technology, AI video production, and compliance-heavy verticals. We're hiring one AI-native engineer who takes detailed briefs and ships real work — websites, automations, API integrations, multi-agent workflows — using Claude Code and the modern AI coding stack to move at triple the speed of a traditional dev.
This is not a prompt-writing job. This is not a "use ChatGPT to help with tasks" job. We need someone who lives in Claude Code, builds subagents and skills files in their sleep, and has shipped enough production work to know what breaks at 2 a.m.
What You'll Actually Build
- Websites and landing pages — Next.js, Vercel, headless CMS, occasional WordPress
- Real estate data pipelines — MLS pulls, agent transaction lookups, HUD parsing, license verification, lead enrichment
- API integrations — FAL, Seedance, Minimax, Supabase, Stripe, Mux, Clerk, Inngest, Resend, Google Drive, Slack
- Multi-agent workflows — Claude Code subagents, MCP servers, n8n flows, background jobs, webhook-driven pipelines
- Internal tools — VA dashboards, QA review interfaces, batch processors, audit logs
- Data enrichment and tagging — structured outputs, vector search, automated categorization
What We Offer
- Starting $1,200–$3,000/month based on demonstrated ability and portfolio
- Full access to our AI stack — Claude Max, Kimi Code, GLM, Minimax, FAL, and more. We pay for all of it.
- Recurring long-term work across multiple brands — this is an ongoing pipeline, not one project
- Direct work with ownership, no middle layers
- Top performer moves into a lead role managing VAs and junior builders
How to Apply — Read This Carefully
Most applicants to jobs like this paste AI-generated answers. We can tell. The candidates we hire answer like people who have actually done the work and have scars to prove it.
To apply, answer the following. Short answers are fine — we'd rather read three honest sentences than three paragraphs of polished nothing. Answer in your own words, in one sitting, without running your response through an LLM to "clean it up." If your answer reads like ChatGPT wrote it, we'll assume ChatGPT did.
1. What's the most annoying thing about Claude Code that you've had to work around? Be specific. What did you do about it?
2. Describe the last subagent, skill file, or slash command you built. What problem did it solve, and what would you do differently if you built it again?
3. You're given a brief: "Pull every active MLS listing for a specific agent, enrich each with public HUD transaction history, and dump the result as a structured JSON file into Supabase storage, triggered by a webhook from a form submission." Give us your rough architecture in 5–10 bullets. Not the code. Just how you'd stitch it together and what would be the hardest part.
4. Name one MCP server you've actually used in production (not just installed to try it), and one you wish existed but doesn't.
5. Which do you reach for first for background jobs — Inngest, n8n, BullMQ, Temporal, or plain cron — and why? No wrong answer, we want to hear your reasoning.
6. Tell us about a project where something broke in production. What happened, what did you do, and what did you learn? One paragraph max.
7. Send portfolio links: GitHub repos, live sites, Looms of tools you built, or anything else that proves you ship. Minimum three links.
8. What's your monthly rate expectation within the $1,200–$3,000 range, and how many hours per week can you commit?
9. One-line answer: what time zone are you in, and what hours overlap with U.S. Eastern?
We read every application that answers all nine honestly. We delete the ones that don't. Copy-pasted generic cover letters, AI-polished "thrilled to apply for this exciting opportunity" openings, or answers that dodge the specifics will not get a response.
Bonus (Not Required, But Noticed)
- Shipped work with Kimi K2.5, GLM, Minimax, or other non-Anthropic coding models
- Experience with FFmpeg, video pipelines, or media processing
- Real estate, mortgage, or regulated-industry project history
- GitHub with visible commit history over the last 12 months
- Supabase, Vercel, Cloudflare Workers, or Runpod experience