Building AI-native systems where the model is the orchestration layer, not just a feature.
Career
2025
2025
Full Stack & AI Engineer
2025 – Present
Building AI-first, agent-driven tooling for the restaurant industry — shipping reliable multi-agent and large language model features into production workflows.
2024
2024
Full Stack Consultant
Dec 2024 – May 2025
Code audit for a Colombian real estate platform — identified critical architectural flaws blocking production. Implemented key fixes, redesigned front-end components, and deployed to Firebase.
Full Stack Engineer
Jan 2024 – Jun 2024
Built a dual-application real estate platform — property browsing + management dashboard. Engineered reusable components adopted across 3 projects (−50% dev effort) and re-engineered critical modules for a 68% performance gain.
2023–25
2023–25
Full Stack Developer
Aug 2023 – Nov 2025
Migrated a NestJS monolith to a microservices architecture with horizontal scaling and rolling deployments on AWS. Built event-driven systems with BullMQ, Redis, and WebSockets — including a high-throughput call engine (10K+ calls/day) and a queue-based mass messaging platform. Integrated Twilio for SMS/voice and AI models for email parsing and message generation.
About
I'm a Full Stack & AI Engineer from Pereira, Colombia, with a focus on building systems where AI is the intelligence layer — not just a wrapper around a model.
My work spans event-driven backend architectures and microservices migrations to integrating LLMs into production workflows. I'm drawn to problems at the intersection of distributed systems and machine intelligence. The frontend side comes naturally too — building interfaces that feel polished and easy to use is just as important to me as the system underneath.
I paused my Systems & Computer Engineering degree at Universidad Tecnológica de Pereira to build a more concrete path — one focused on what’s new and what the world actually needs. My direction is to become an agentic software engineer, going deeper into LLMs, fine-tuning, and model design while shipping world-class products in production.
Expertise
Work
AI-native operating system for Colombian small businesses
Building Beela — an AI-native OS for Colombian small businesses. Scheduling, services, payments, POS, customer messaging, and analytics in one product, with multi-agent orchestration on top. Owners book through WhatsApp or voice, clients through a public site every business gets out of the box, and a daily-brief agent surfaces what is worth attention each morning. Multi-tenant, role-configurable, real-time, bilingual (es-CO / en).
Tree-structured distributed memory for AI agents
A framework built on distributed database concepts and tree data structures — giving AI coding agents persistent, structured memory across sessions and environments. Agents share decisions, context, and knowledge through a connected graph: every node is a memory, every edge is a relationship.
Full Stack bootcamp — Pereira, Colombia
Co-directed a 6-week intensive Full Stack Web Development Bootcamp with faculty from Universidad Tecnológica de Pereira. Rebuilt the program around market-ready technologies, modern engineering practices, and AI developer tooling — preparing students to enter the industry production-ready.
FAQ
Adrian Gaitan — also written Adrian F. Gaitan or Adrian Gaitan Londoño — is a Full Stack & AI Engineer based in Pereira, Colombia. He builds AI-native products and production-grade AI agents: multi-agent orchestration, event-driven backends, and LLMs running in real workflows. He is not the visual artist of the same name.
He designs and ships AI-native software where the model is the orchestration layer — not a bolt-on feature. His work spans multi-agent systems, microservices, event-driven architectures, and full-stack web apps in TypeScript, Node.js, NestJS, React, Next.js and Astro.
Yes. Adrian is open to full-time roles, contract engineering, technical leadership, and collaboration — remote, worldwide. The fastest way to reach him is by email at [email protected].
He is a Full Stack & AI Engineer at KitchenSync, an AI-first back-office OS for restaurants, where he ships multi-agent and LLM features into production. He also builds Beela, SwiftControl, and an Agent Memory Layer framework for persistent AI-agent memory.
Pereira, Risaralda, Colombia (UTC−5). He works remotely with teams around the world and speaks Spanish (native) and English (professional).
Email [email protected], or find him on GitHub (@adriancho91s) and LinkedIn (in/adriancho91s). You can also book a call directly from this site.
Get in touch
Open to full-time roles, freelance, and interesting side projects. If you have a problem worth solving, let's talk.