Open to opportunities

Francisco Barrera

@franciscobarrera

I design production LLM-powered systems—automating conversational and voice workflows with robust infrastructure and APIs.

Ecuador

Message

What I'm looking for

I want to build and deploy production LLM systems that automate real business workflows—combining conversational AI, voice AI, and reliable infrastructure—where I can own end-to-end architecture, integrations, and measurable outcomes.

I’m an AI Systems Engineer with 7+ years of experience across IT infrastructure, software development, and systems architecture. I specialize in designing and deploying production AI systems powered by Large Language Models (LLMs), with a strong focus on end-to-end automation that connects conversational AI, voice AI, and intelligent workflows to real business operations.

In my current role as an LLM Engineer & AI Systems Architect (remote, Ecuador/Colombia/Peru), I built a production chat assistant on Gemini 2.5 Flash, deployed as a systemd service behind an Nginx reverse proxy. I also architected automated outbound calling with VAPI (GPT-4o Mini + ElevenLabs TTS), including geographic routing, Google Calendar scheduling, WhatsApp confirmations via Twilio, and CRM updates through orchestrated Make.com scenarios. Alongside this, I designed a 25-node n8n lead qualification pipeline (Meta webhook ingestion → normalization → duplicate detection → Gemini scoring → regulatory positioning → automated email for qualified leads) and created a Python/ReportLab PDF quote engine, supported by a production stack I manage end-to-end (Ubuntu VPS, Docker, Cloudflare SSL, Supabase, and self-hosted n8n).

Experience

Work history, roles, and key accomplishments

Current

LLM Engineer & AI Systems Architect

Current

Arquitectura Financiera

Jan 2024 - Present (2 years 3 months)

Designed and deployed production LLM systems for financial operations across three countries, including a Gemini-based conversational assistant and an automated outbound call center with WhatsApp confirmations. Built a 25-node n8n lead pipeline with 0–100 qualification scoring and automated email notifications for 60+ qualified leads, plus a Python/ReportLab PDF quote engine for 10 products.

Gemini API Node VAPI N8n Make.Com Python Nginx Docker

IT Systems Intern

Prodegel S.A.

Aug 2017 - Apr 2018 (8 months)

Maintained and upgraded main server infrastructure, including firewall configuration, and supported hardware/software maintenance cycles and internal process monitoring.

System Administration Maintenance Software Maintenance Linux Administration Technical Support