AI Integration Engineer (LATAM)

AI Integration EngineerWe built 3 AI agents in 48 hours. You'll help us build the next 10.Who We AreEightX is a growth economics consultancy that serves eCommerce, DTC, and CPG brands doing $2M–$80M in revenue. We act as fractional CFOs and growth advisors — sitting inside our clients' businesses, running financial models, managing ad performance, and tracking every deliverable.We're now building an internal AI agent system that automates the work our team does every day: reading call transcripts, updating financial models, monitoring ad performance, tracking client deliverables. Three agents are already live in production on a self-hosted VPS, wired to Slack, Fireflies, and Meta. You'll be the engineer who builds everything next.The RoleYou'll work directly with the founder to build, ship, and maintain the AI agent infrastructure. This is not a research or architecture role — it's a build role. We move extremely fast. You'll be expected to take a spec, pick it up in Claude Code, and have something working the same day.You'll own:Building MCP servers (Node.js) that connect our AI agents to external APIs — Fireflies, Gmail, GoHighLevel CRM, Google Sheets, Meta AdsWriting skills, .md files enabling our teams with OpenClaw, Claude Cowork and LLM APIs.Google Sheets automation — scripting financial model updates, building structured data pipelines from call transcripts to spreadsheetsInfrastructure work — Docker, SSH, container health monitoring, environment variable management, network isolationDebugging fast — reading logs, finding the failure point, shipping the fixWhat You Must HaveNon-negotiableYou use Claude Code (or Cursor) as your primary development environment — not as a novelty, but as your main way of working. If you're not already doing this, this role isn't right for you.You ship today, not next week. If the spec is clear, you build it the same day. That's the bar.Node.js — comfortable with async/await, HTTP servers, API integrations, npm ecosystemDocker — can build images, run containers, read logs, manage environment variables, debug networking issuesREST and GraphQL API integration — reading API docs and implementing against them is bread-and-butter work for youYou must understand how LLMs think, how they ingest information and how to structure information and prompts for maximal effectivenessStrong PreferenceExperience with MCP (Model Context Protocol) server development or any similar tool-server patternGoogle Sheets API or Apps Script — we use Sheets as a core client deliverable formatSlack bot development or webhook integrationsPrior work building automation systems — n8n, Make, Zapier, or custom API glue that you outgrew and replaced with codeComfortable with Linux and SSH — this is not a managed cloud environmentWhat You'll Build FirstIn your first two weeks, you'll:Build out 4 highly functional, secure and intelligent agents. We will show you how. You will deliver.Replicate a couple agents to work client sideBegin enabling our team to use and interact with new agentsThese aren't speculative projects. The specs exist. The VPS is running. You're shipping into production from day one.How We WorkAsync-first. Most communication via Slack. No daily standups.You'll get clear written specs. We don't expect you to figure out what to build — we expect you to build it fast.We use Claude Code heavily. We expect you to use it too. We'll share our patterns, prompts, and learnings.Full-time contractor. Ongoing engagement. This isn't a one-time project.Timezone: LatAm or India preferred. LatAm strongly preferred for real-time overlap with US hours.How to ApplyWe don't want a cover letter. We want to see you work.Send us:A link to something you've built — an API integration, a bot, an automation, a tool server. Doesn't have to be polished. Has to be real.One paragraph on how you use Claude Code or Cursor in your daily workflow. Be specific — not 'I use AI to help me code' but how and for what.Your hourly rate and timezone.If your profile looks right, we'll send you a paid test task: build a working MCP server against a real API from our stack. You have 30 minutes. That's not a typo. We'll pay you for your time regardless of outcome — but if 30 minutes sounds impossible, this role probably isn't the right fit.Short Video Submission – Early ScreeningAs part of our initial screening process, candidates will be asked to submit a short video. This helps us assess communication skills, genuine interest in the role, and basic video/audio setup.Video GuidelinesLength: 2–3 minutesFormat: Phone or webcamPlease address the following in your video:Why are you interested in this position?What are your key strengths related to the role you're applying for