Carlos Salgado

Data Scientist | Machine Learning | AI Enthusiast

About Me

Data Scientist with over 5 years of experience in end-to-end projects, delivering scalable and results-driven solutions for decision-making. Specialist in transforming complex data into actionable insights, optimizing processes, and generating business value. Strong expertise in Sales Forecasting, Computer Vision, Machine Learning, and Business Intelligence, with proficiency in Python, SQL, Docker, Flask, Qlik Sense, Power BI, and Excel. Experienced in agile environments, leading projects from conception to final delivery, ensuring real impact for clients and stakeholders.

My Projects

Sales Demand Forecasting

This repository contains the final exploratory analysis notebook for the sales forecasting project. The goal of this analysis is to understand sales variability and identify significant patterns that can guide marketing strategies and resource optimization.

View on GitHub

Credit Score Prediction ML Models

This repository contains the final exploratory analysis notebook for the Credit Score Prediction project. The goal of this analysis is to predict the score from a Credit.

View on GitHub

E-commerce Backend Project

This project is a backend for an e-commerce system, implemented with SQLite as the database. It includes basic functionalities for managing orders, products and the shopping cart. The project consists of Python scripts for manipulating data in the database and automated tests to ensure code integrity.

View on GitHub

Health Plan Costs

This project aims to develop a methodology for predicting health plan costs for beneficiaries. It includes data exploration, preprocessing, and machine learning model evaluation. The analysis considered factors such as age, muscle mass, children, gender, smoking, and region. Various regression models were tested, with Random Forest achieving the best performance (R² = 0.90).

View on GitHub

Open/Close Bottle with Computer Vision

This project is a computer vision system that detects whether an object is open or closed. It utilizes image processing techniques and machine learning models to analyze visual data. The implementation includes dataset preprocessing, model training, and evaluation to ensure accurate classification.

View on GitHub