Skip to the content.
Carolina Rios-Trujillo

💻 My Code Projects

Here you can explore some of the projects I have developed. Each project includes a brief overview and a link to its GitHub repository (or Live Demo).

Generative AI

RAG-Powered Study Assistant & Test Maker

RAG system that generates practice quizzes strictly from course materials (no hallucinations). Features reranking, metadata filtering by lecture, and a fully deployed web interface.

Tech Stack: Python · GPT-4o-mini · ChromaDB · HuggingFace · Docker · Render · Flask

Live Demo View Project
Deep Learning

Connect 4 Deep Learning Agent

End-to-end Connect 4 AI using MCTS self-play for data generation and CNN/Transformer models to learn move-selection policies. Deployed into a live gameplay interface.

Tech Stack: Python · TensorFlow/Keras · CNN · Transformer · MCTS · AWS · Docker

Live Demo

Demo credentials — Username: dan · Password: Optimization1234

Recommendation

WanderLens: A Camera-Roll Recommendation System

Developed an end-to-end recommender that infers user travel preferences from personal photo libraries using LDA and semantic similarity, generating personalized travel activity suggestions.

Tech Stack: NLP · LDA · TF-IDF · Embeddings · Cosine Similarity

🔧 Repo coming soon

Predictive ML

Formula One Outcomes Prediction

ML project using telemetry, weather, and qualifying data (2022–2024) to predict driver probabilities. Built regression and classification models and simulated weather scenarios for race strategy insights.

Tech Stack: Python · Scikit-learn · XGBoost · Pandas · Feature Engineering · Model Evaluation · Data Visualization

View Project
Python · OOP

Business Valuation Estimator

A Python program that uses OOP to estimate business value using CAGR and DCF, with user input and validation.

Tech Stack: Python · OOP · CLI Input · Modular Design

View Project
Recommendation

Applied Data Science Projects Course

Explored customer order data to identify demand trends and improve business strategies. Built recommendation models (Rank-Based, Collaborative Filtering, Matrix Factorization) to enhance user experience.

Tech Stack: Python · Pandas · Scikit-learn · Matrix Factorization

View Project
Classification

Predicting Credit Card Churn

A classification model that predicts credit card churn in the banking industry.

Tech Stack: R (data.table, dplyr, ggplot, xgboost, Metrics)

View Project