LLM + RAG for public brazilian companies
This project combines Large Language Models (LLM) with Retrieval-Augmented Generation (RAG) to efficiently answer queries about public Brazilian companies. I’ve used PostgreSQL to …
Professional Summary
M.S. in Computer Science from IME-USP, focused on Reinforcement Learning. Founder of 2 companies, 10+ years of experience
working with large-scale databases and building end-to-end ML pipelines. Kaggle competitor and Scikit-learn contributor.
Education
MSc. Computer Science
2021-03-01
2025-09-01
University of Sao Paulo (IME-USP)
BS Computer Engineering
2008-01-01
2012-12-01
Faculdade de Informatica e Administracao Paulista (FIAP)
Interests
Reinforcement Learning
Deep Reinforcement Learning
Machine Learning
Large Language Models
Stochastic Shortest Paths
Projects
Inverse Reinforcement Learning for epidemiology
There is a big challenge into designing a manual reward function to optimize sequential decision making during epidemics/pandemics. Thus, I proposed to use Inverse Reinforcement …
Providentia Forecasting
Providentia Forecasting is a service to do demand forecasting focused on the industry sector. The general idea is that any person could generate forecasts based on multiple …
Recent Posts
Finding Meaning in Sparse Rewards
Some problems in life are inherently sparse. You can work hard, make many decisions, and still receive little or no feedback. Imagine working on a large research project you …
Ranking Counter-Strike 2 teams using Bradley-Terry model
Introduction When I was younger, I enjoyed playing Counter-Strike (CS) with my friends. I started with version 1.5, then moved on to 1.6, and later played CS:GO. Nowadays, I’m more …
Data Science for Impeachment Votes Prediction
1. Introduction It is not new that Data Science is applied to the political context, but we can say that the 2008 US presidential elections were a watershed for Data Science in …
Data Science para previsão da votação do Impeachment
1. Introdução Não é de hoje que Data Science é aplicado ao contexto político, mas podemos dizer que as eleições presidenciais dos EUA de 2008 foram um divisores de água para Data …
Faça coisas que não escalam (Do Things That Don’t Scale)
Artigo original em http://paulgraham.com/ds.html Por Paul Graham Table of Contents Recrutar Frágil Deliciar Experiência Fogo Meraki Consultar Manual Grande Vetor Notas Traduzido …
Conselho Para Os Ambiciosos Jovens De 19 Anos
Texto original disponível em: https://blog.samaltman.com/advice-for-ambitious-19-year-olds Por Sam Altman Traduzido por Gustavo De Mari Pereira “Eu sou um ambicioso jovem de 19 …