Managing Research Data

Organizing large-scale research datasets requires a disciplined approach to versioning and project structure. The data_cancer_svm project focuses on streamlining the preparation and management of diagnostic data for machine learning models. A significant step in this process involves ensuring that all necessary data files are systematically ingested into the

Building a neural network from scratch often feels like magic until you see the numbers behind the curtain. The redes_neuronales_numeros_mnist project explores the fundamentals of image classification using the classic MNIST dataset, providing a clear window into how machines learn to interpret handwritten digits.

The Anatomy of an Image Classifier

At its core, a neural network is

Building machine learning models for sensitive classification tasks requires a delicate balance between algorithmic precision and code maintainability. In the data_cancer_svm project, recent updates to the application logic demonstrate how modularizing data processing steps can simplify complex SVM workflows.

The Complexity of Classification

When working with Support Vector Machines (SVM) in

Introduction

In the telco-reg-app project, our focus has shifted toward streamlining how we ingest and process regulatory telecommunications data. To ensure our data layer remains performant, we have been refactoring our ingestion modules to better leverage vectorized operations for large-scale datasets.

The Challenge

Previously, processing large volumes of telecommunications records

Introduction

The aws-rekognition project serves as a foundational bridge for integrating computer vision capabilities into web applications. By leveraging cloud-native AI services, we can process and analyze image data directly within our application flow.

The Challenge

Implementing computer vision in a traditional web stack often presents hurdles, including:

• Handling binary image

Maintenance isn't just about adding new features; it is equally about removing the dead weight that slows down your development cycle. In the joseaholgado/ANNs-Zalando project, a repository focused on implementing Artificial Neural Networks using TensorFlow and Keras, we recently performed a cleanup to improve codebase clarity.

The Cost of Clutter

When working on machine learning

Improving Data Handling Practices

Maintaining a clean and predictable file structure is a cornerstone of reproducible research. Recently, in the ANNs-Zalando project, a small but significant adjustment was made to how serialized model data is stored. By explicitly updating the file extension of the dataset artifacts, the project improves clarity for developers and automated tooling alike.

In the data_cancer_svm project, I recently undertook a cleanup task focused on improving the clarity of our serialized model artifacts. While seemingly minor, tracking metadata in serialized files is critical for reproducibility in machine learning workflows.

The Problem: Obscure Artifacts

When we save pre-processing objects—like scalers or encoders—to disk, the naming convention often

Building Autonomous Workflows

Modern application development is shifting from static request-response cycles to dynamic, agentic workflows. In our work with the agente-langgraph project, we are exploring how to orchestrate intelligent agents that can reason, plan, and execute tasks across complex multi-step processes using the LangChain ecosystem.

The Shift to State-Based Orchestration