Data Science & AI Undergraduate | Generative AI & Data Engineering Intern 📍 Brasília, Brazil | 📧 | LinkedIn

I build production-ready AI and data engineering solutions, from scalable LLM applications and retrieval-augmented generation (RAG) pipelines to robust ETL workflows with SAP and modern data stacks. I transform complex data challenges into efficient systems that drive business impact.

LLM-powered PDF parsing • Streamlit dashboard • SQLite persistence • Batch processing

Python LLM Streamlit Polars SQLite PDF Processing

• Engineered a modular system for large-scale legal PDF parsing and structured data extraction
• Optimized batch processing to handle documents up to 200MB with minimal memory footprint
• Implemented schema validation and auto-repair mechanisms to ensure data integrity
• Delivered a production-ready UI enabling real-time analytics and manual review workflows

RAG pipeline • Domain-specific intent detection • Streamlit UI

RAG Sentence-Transformers Streamlit FAISS SAP BODS

• Developed a custom chatbot tailored for SAP BODS users, overcoming generic LLM limitations
• Integrated semantic search and intent classification for precise, domain-aware support
• Enabled fluent, natural Portuguese interactions aligned with SAP ecosystem documentation

XGBoost • Model Explainability • Streamlit UI

XGBoost SHAP Streamlit Scikit-learn Pandas

• Delivered highly accurate delivery time predictions using structured business features
• Applied SHAP for interpretable models, providing actionable insights to stakeholders
• Built a user-friendly interface for real-time predictions and scenario analysis

Sentiment Analysis • Text Generation • Multi-modal Applications

Gemini AI Polars Sentiment Analysis Text Generation

• Built a sentiment-aware Food Review Reply Bot generating empathetic responses to customer feedback
• Created a persuasive Dish Description Generator to boost engagement in food delivery apps
• Designed efficient data pipelines with Polars for scalable multi-modal applications

Hybrid Neural Network • Collaborative Filtering • Deep Learning

PyTorch Neural Networks Collaborative Filtering Hugging Face

• Designed a hybrid recommendation engine combining deep learning with collaborative filtering
• Developed scalable workflows in PyTorch for personalized book suggestions
• Documented the entire process through a comprehensive Medium tutorial

Interactive Dashboards • Healthcare Analytics • Automated Reporting

PyGWalker Polars Streamlit Plotly Gradio

• Delivered instant exploratory data analysis with an interactive Fast EDA dashboard
• Developed a visual analytics platform for Brazilian live births (SINASC 2023 data)
• Authored an interactive tutorial for PyGWalker enabling fast, code-light EDA

Programming & Query Languages: Python SQL C Bash

AI & Machine Learning: PyTorch Transformers Hugging Face XGBoost Scikit-learn LLMs RAG Embeddings

Data Engineering & Processing: SAP Data Services (BODS) Pentaho Polars Pandas SQL Databases ETL/ELT

Data Visualization & Apps: Streamlit Gradio Plotly PyGWalker Power BI

Tools & Platforms: Git Docker Linux Hugging Face Spaces Google Colab

Certifications: Oracle Generative AI Professional Oracle Data Science Professional

Data Migration Intern — First Decision (Apr 2025 – Present)

• Delivered complex ETL migrations for enterprise clients, handling end-to-end data workflows
• Developed and optimized pipelines leveraging SAP BODS, Pentaho, and SQL databases
• Gained hands-on experience managing data migrations for large-scale market players

B.Sc. Data Science & Artificial Intelligence — IESB (2023–2026) Core Courses: Machine Learning, Data Mining, Big Data, Statistical Modeling, Deep Learning