Skip to content
View aletbm's full-sized avatar
😁
I don't have bugs, I have stochastic features
😁
I don't have bugs, I have stochastic features

Block or report aletbm

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Please don't include any personal information such as legal names or email addresses. Maximum 250 characters, markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
aletbm/README.md

👋 Hola, soy Alexander Daniel Ríos

💻 Data Scientist | Machine Learning Engineer
Con más de 4 años de experiencia en proyectos prácticos de alto impacto.
Especializado en MLOps, Deep Learning (NLP & CV) y despliegue de modelos en producción.


🛠️ Tech Stack

Python TensorFlow Keras Scikit-learn Docker FastAPI Prefect Terraform GCP spaCy MLflow Evidently Matplotlib Seaborn Pandas Dask LangChain SmolAgent LlamaIndex Gemini Flask LangGraph Transformers Hugging Face YOLO


🚀 Proyectos Destacados

  • Pipeline end-to-end con MLflow, Prefect, FastAPI, Docker, Terraform, GCP, Evidently.
  • Accuracy 0.89 | F1 0.89 con KNN.
  • Despliegue en Google Cloud Run + monitoreo de drift + alertas en Slack.
    👉 Análisis | API | Reporte

  • Modelos de Deep Learning (U-Net, CNN) para clasificación y segmentación.
  • ROC AUC 0.99 | F1 0.97 (clasificación)
  • ROC AUC 0.98 | BinaryIoU 0.94 (segmentación)
  • Despliegue como servicio REST + app interactiva en Streamlit.
    👉 Análisis | App

🔹 Clasificación de tweets sobre desastres (Jul 2024 – Ago 2024)

  • Modelo LSTM + BERT sobre 10k tweets.
  • F1-score 0.84 en validación cruzada.
  • App en Streamlit para clasificación en tiempo real.
    👉 Análisis | App

  • Forecasting multiseries temporales con XGBoost y CatBoost.
  • Top 4 en la competencia oficial (RMSE 8.99).
    👉 Notebook

📚 Artículos y Publicaciones


🎓 Educación & Formación

🎓 Ingeniería en Electrónica – UTN, Buenos Aires (2014 – 2026)
🎓 Técnico Universitario en Electrónica – UTN, Buenos Aires (2014 – 2018)

📘 Machine Learning Zoomcamp – DataTalks.Club (Sep 2024 – Ene 2025)
📘 MLOps Zoomcamp – DataTalks.Club (May 2025 – Ago 2025)
📘 AI Agents – Hugging Face (May 2025 – Jun 2025)


🌍 Idiomas

  • Español: Nativo
  • Inglés: B1 Intermedio (EF SET)

📊 Estadísticas de GitHub

Estadísticas GitHub
Lenguajes más usados GitHub Streak


📫 Contacto

📍 Buenos Aires, Argentina
📧 alexanderdaniel_rios@hotmail.com
🔗 Linktree
💼 LinkedIn
💻 GitHub

Pinned Loading

  1. Blood_Cell_Cancer_Prediction Blood_Cell_Cancer_Prediction Public

    Capstone project for the DataTalks.Club ML Zoomcamp focused on detecting acute lymphoblastic leukemia (ALL) using image classification and segmentation. The project includes model training and an i…

    Jupyter Notebook 3

  2. Intrusion_Detection_E2E_MLPipeline Intrusion_Detection_E2E_MLPipeline Public

    End-to-end MLOps pipeline for intrusion detection using MLflow, Prefect, FastAPI, Docker, Terraform, and GCP. Includes model training, deployment, monitoring, and CI/CD.

    HTML 7 1

  3. NLP_with_Disaster_Tweets NLP_with_Disaster_Tweets Public

    Capstone project using NLP to detect if tweets describe real disasters. Built on 10,000 labeled tweets to support emergency response with AI. Includes a Streamlit app for real-time tweet classifica…

    Jupyter Notebook 1

  4. Hazardous_Asteroid_Classification Hazardous_Asteroid_Classification Public

    ☄️ Machine Learning project for classifying potentially hazardous asteroids using NASA JPL data 🌍. Includes full preprocessing, orbital analysis, model deployment, and an interactive Streamlit app 🚀.

    Jupyter Notebook

  5. Retail_Demand_Forecast_MLZoomCamp_Competition_2024 Retail_Demand_Forecast_MLZoomCamp_Competition_2024 Public

    🛒 Forecasting retail demand using time series data. Developed for DataTalks.Club's ML Zoomcamp 2024 — proudly achieved 4th place in the official competition! 📊🚀

    Jupyter Notebook 1

  6. ARFX__PDS_aplicado_en_audio ARFX__PDS_aplicado_en_audio Public

    Procesamiento digital de señales aplicado en audio con Python - efectos con líneas de retardo, saturación, moduladores de amplitud, reverberación, vocoders, filtros y método de reconstrucción de me…

    Python