Arturo Vaine

Built a back-office platform (MERN) with MCP for task management and reporting across a 60+ person team. Deployed ETL pipelines and a RAG system consolidating 2,000+ reports into a searchable knowledge base.

Migrated spreadsheet-based and legacy workflows into a maintainable web application. Translated existing business rules and operational processes into application logic.

Founding Engineer: set up the company's GitHub org and bootstrapped 7 projects; AI strategy for SMBs. Built an AI engine for construction projects, analyzing PDFs and images to identify bid risks.

Built MCP and audiovisual pipelines using Python, Zapier, webhooks, and FFmpeg to process TV episodes. Achieved 80% time savings by automating docs processing with ETLs and vision-language models.

Deployed Medallion Architecture in BigQuery and APIs (SQLAlchemy, FastAPI) for data ingestion. Integrated data pipelines and analytics using GCP, Retool, and Metabase.

Built digital menu from scratch (React) to open new revenue channel for more than 300 customers. Developed integration and applications: portfolio management, broadcast messages, etc. (NoSQL, Firebase).

Delivered data through ETL / BI of R$ 160MM+ annual GMV and other KPIs (MySQL, PowerBI, Streamlit). Developed MERN stack for Product Information Manager (PIM) of portfolio enrichment w/ GenAI (180k+ SKUs).

Delivered key features to enhance APIs / back-end of National Power Plant Monitoring System. Refactored and modernized legacy Express.js codebase by transitioning to a multi-layer architecture REST API.

Developed data collection solution of technical form with ~350 inputs (HTML, Vanilla JavaScript, CSS; SQLite). Delivered data analysis process to check inconsistencies and generate ~900 reports (Python-Pandas).

Bachelor's degree in Mechanical Engineering from the Federal University of Technology – Paraná, completed in 2016.

Visiting student for two semesters in Mechanical Engineering, approximately 1000 hours, as part of the 2013 cohort.