AI & Automation Developer | Building intelligent workflows & LLM-powered applications
I craft automation solutions that scale. From RAG systems that actually know their stuff to n8n workflows that handle the repetitive, I focus on the intersection of AI and productivity.
- Workflow Automation: n8n, Make.com — integrating APIs, automating repetitive tasks
- RAG Systems: Building knowledge bases that prevent AI hallucinations (think: verified recipes, trusted sources)
- LLM Integration: OpenAI, prompt engineering, AI-powered features
- Bot Development: Telegram bots, content processing pipelines
- Data Processing: FFmpeg integration, content workflows, social media automation
- Frontend: React, Angular, Vue | Tailwind CSS, vanilla JS
- Backend: Node.js, Express, Python/Django
- Databases: PostgreSQL, MongoDB, SQLite, MySQL
- Testing: Cypress, Playwright, Vitest
Telegram bot powered by AI with a twist — uses Supabase + RAG to prevent hallucinations. Recipes come from a verified corpus, not LLM dreams.
End-to-end automation: data ingestion → processing → storage → API. Built with n8n, Supabase, and OpenAI.
Workflow automation for media processing: video transcoding (FFmpeg), metadata extraction, social media integration.
const itziarStack = {
automation: ["n8n", "Make.com"],
backend: ["Node.js", "Express", "Python", "Django"],
databases: ["PostgreSQL", "MongoDB", "SQLite", "MySQL"],
ai: ["OpenAI API", "RAG", "LLMs", "Prompt Engineering"],
frontend: ["React", "Angular", "Vue", "Tailwind CSS"],
tools: ["Git", "FFmpeg", "Telegram API", "Supabase"],
testing: ["Cypress", "Playwright", "Vitest"],
}- 🔨 Building automation side projects with AI
- 📚 Deepening RAG knowledge & LLM workflows
- 🎯 Exploring agentic AI patterns
⭐️ @itziarZG