This project automates a daily stock trading strategy using GRU-predicted highs and Alpaca API. It fetches historical stock data, predicts the next day's high, and places buy/sell orders on the market open and target price.
This is the Angular frontend for the Otero Ediciones landing page. It displays a dynamic book catalog and a brief history of the editorial.
A tactical turn-based board game where two players duel 1v1 using movement cards instead of fixed rules. Each turn, cards are exchanged, forcing constant adaptation as you maneuver your pawns to capture the enemy King or invade their Base before they outplay you.
Free online audio tools built with Go, templ, HTMX, AlpineJS and Tailwind.
Relax is a lightweight, high-performance Deep Learning library written from scratch in Zig. It features a dynamic reverse-mode automatic differentiation engine (Autograd), N-dimensional tensors with broadcasting, and a high-level, Keras-style API for building and training neural networks.
A lightweight school management platform built in pure Go, designed for managing users, classes, assignments, and submissions. The project focuses on simplicity, performance, and clean architecture without external frameworks.
Gorillaz is a machine learning library written in Go. It provides tools for building and evaluating various machine learning models, including regression, classification, and utilities for model persistence and evaluation metrics.
Classifier model to detect skin cancer based on mole images. The model was used in the app MedAL
StockPulse is a university project that analyzes the sentiment of financial news related to publicly traded companies. It combines a lightweight backend architecture (C# API + FastAPI NLP service) with a responsive React frontend.
Grizzly is a DataFrame library for Go, designed to harness the full power of GoRoutines for handling large datasets efficiently. Its core aim is to provide an easy-to-use, yet robust, solution for data manipulation while maximizing the computational capabilities of modern machines through parallelized task execution.
This project aims to predict house prices based on various features using different Machine Learning models. The objective is to identify the best-performing model by evaluating multiple regression algorithms and assessing their performance using common error metrics.
Team project analysing MSFT stock time series. LSMT approach using Keras framework.