Full-Stack Engineer (TypeScript & AI Agents)

📍Medellín, Colombia (Hybrid — in-person days negotiable)
💼Full-time · 4–6M COP/month

Join Medallo to build AI-forward products end to end. You'll work across the stack with TypeScript, React, Next.js, Supabase, and Tailwind — and go deep on agent workflows: Hermes and OpenClaw-style agents, MCP servers, skills, and LLM integrations. This is an engineering role built around modern AI tooling; we use Cursor daily and expect you to ship faster with AI-assisted development, whether or not you've used Cursor or Claude Code before.

Responsibilities

  • Build and maintain full-stack features with TypeScript, React, Next.js, Supabase, and Tailwind CSS
  • Design and implement AI agent integrations — tool calling, memory, and orchestration patterns
  • Develop and connect MCP servers and agent skills for production workflows
  • Write clean, typed code across frontend, API routes, server actions, and database layers
  • Collaborate on architecture for LLM-powered features: prompts, structured outputs, and error handling
  • Use AI-assisted development tools (especially Cursor) as a core part of your daily workflow
  • Participate in code reviews, debugging, and iterative product delivery

Requirements

  • 2+ years of professional software development experience
  • Strong proficiency in TypeScript and modern JavaScript
  • Hands-on experience with React and Next.js (App Router preferred)
  • Experience with Supabase or similar Postgres-backed BaaS (auth, RLS, Edge Functions)
  • Comfortable styling with Tailwind CSS
  • Solid understanding of full-stack patterns: REST/RPC, server-side rendering, and database design
  • Genuine interest in AI-assisted development and willingness to adopt Cursor as your primary IDE
  • Good communication skills; B2+ English for technical documentation and async collaboration
  • Based in or willing to relocate to Medellín for a hybrid arrangement

Nice to Have

  • Prior experience with Cursor, Claude Code, or similar AI coding assistants
  • Exposure to AI agents (Hermes, OpenClaw, LangGraph, or custom agent frameworks)
  • Experience building or consuming MCP (Model Context Protocol) servers
  • Familiarity with LLM APIs (OpenAI, Anthropic, etc.) and structured output patterns
  • Experience with agent skills, tool definitions, or prompt engineering in production
  • Knowledge of RBAC, OAuth flows, or webhook-driven integrations
  • Contributions to open-source or personal projects showcasing AI + web stack work