Olist Open Source Projects

Data analysis about Brazilian e-commerce business Olist

In this analysis, our group looked at data from the years 2016-2018 of 100,000 orders placed on the Brazilian eCommerce platform, Olist. We used the Random Forest machine learning algorithm to create a web app that predicts customer review scores using the most highly correlated features from the data as user inputs.

SQL queries from a Kaggle notebook I wrote to showcase some of the capabilities of SQL for data analysis.

A Comprehensive Time Series Analysis for Dynamic Sales Forecasting

Projeto final da disciplina Estatística I do curso Santander Coders em parceria com a AdaTech ministrada pelo professor Damodara Barbosa.

Supply chain delay predictor using the Brazilian Olist e-commerce dataset — SQL-based EDA, binary classification ML model, and interactive dashboard to identify and predict late deliveries by state, seller, and shipping cost.

Olist Brazil - Sales Insights and Machine Learning Predictions 📊🤖 Developed sales dashboards and machine learning models to analyze customer behavior and product performance for Olist, a Brazilian e-commerce platform. Utilized Power BI for visualization and multiple machine learning algorithms to predict customer satisfaction. 🔍📈

Esse projeto utiliza dados da Olist para importar, tratar e analisar quatro bases de dados com o Power Query no Power BI. O objetivo é fornecer insights valiosos sobre vendas e comportamento do cliente nos principais marketplaces do Brasil.

Repository from the project "Genius" created at the Megahack 2.0 Hackaton by shawee

MCP server for Olist ERP (Tiny) API v3 — 168 tools auto-generated from OpenAPI spec. Docker + SSE/stdio. Multi-tenant.

Pipeline de ETL com Arquitetura de Medalhao (Bronze/Silver/Gold) para analise do e-commerce brasileiro (Olist), com visualizacao em Power BI. Disciplina de Sistemas de Banco de Dados 2 (Eng. de Software, UnB).

🔷 Business Insights & Executive Report for E-Commerce Dataset 🔷 Analyzed sales, profit, delivery performance, and customer behavior across multiple Brazilian regions. Used DAX, data modeling, and visualization techniques for dynamic, interactive dashboards. Identified key growth areas, top-performing categories, and optimized payment insights.

SDK não oficial em TypeScript para a API v2 do TinyERP (Olist)

Ecommerce Analytics Pipeline

Pipeline de dados com Arquitetura Medalhao (Landing, Bronze, Silver, Gold) sobre o dataset Olist usando Databricks e Delta Lake

Hierarchical Bayesian A/B testing on the Olist marketplace dataset. DuckDB SQL pipeline + PyMC; DiD design flips the sign of the naive on-time-delivery result.

Customer segmentation & delivery performance analysis using SQL + Power BI

End-to-end analysis of Olist e-commerce data, including customer segmentation and testing the impact of free-shipping vouchers.