Mohamed Toumi

Responsible for Data & Analytics and Machine Learning projects within ACCIONA Energía, I am in charge of developing and deploying advanced machine learning models to optimize electricity market forecasting, thus contributing to strategic short-term and long-term decision-making. My role encompasses the design of complete solutions, from model development to production deployment, ensuring scalability and operational efficiency:

- Development of electricity forecasting models (day-ahead spot price, tension days, long-term spot prices and fast reserves), based on time series and advanced algorithms (Scikit-Learn, Prophet, TiDE, TFT).

- Deployment of scalable solutions, integrating containerization (Docker) and ML pipelines (acquisition, feature engineering, training, prediction) with MLFlow and CI/CD (GitHub Actions) for reliable automation.

- Creation of a robust data architecture, with documentation and collaborative tools (Confluence, Jira, Github, Docker Hub).

- Impact: significant improvement of models, particularly for the day-ahead spot price (performance superior to Montel, Kpler commercial forecasts)

Technical Stack:

- Python, Java

- SQL

- Docker, Airflow, Kubernetes

- MLFlow

- Git/Github

- REST API

- Jira/Confluence

Skills: MLOps · Machine Learning · Data Science · Time Series Forecasting · Energy, Python · SQL · Data Engineering · Continuous Integration and Continuous Delivery (CI/CD) · Git · Java · REST APIs · Docker · Airflow · MLflow

As a freelance Machine Learning Engineer / MLOps / Data Scientist, I develop advanced machine learning and deep learning algorithms and models. I also deploy scalable solutions, while ensuring their automation and maintenance, both in on-premise and cloud environments.

- Design and development of machine learning and deep learning algorithms for various applications, using frameworks like Scikit-Learn, TensorFlow, PyTorch, etc.

- Deployment of data science models with complete ML pipelines, integrating automation solutions to ensure reliable and reproducible deployment (CI/CD, Docker, Kubernetes).

- Development of MLOps infrastructures to ensure effective model lifecycle management, including performance monitoring, version management, and model updates in production.

- Work on Cloud projects (AWS, GCP, Azure) for the deployment and automation of large-scale ML workflows.

- Close collaboration with data scientists, engineers, and DevOps teams to optimize solutions and ensure the scalability of models in production.

Technical Stack:

- Machine Learning: Scikit-Learn, TensorFlow, PyTorch

- MLOps: Docker, Kubernetes, CI/CD, MLflow

- Cloud Platforms: AWS, GCP, Azure

- Programming Languages: Python, SQL

- Data Engineering: Apache Airflow, Kafka, Spark

Skills: SQL · Computer Vision · PySpark · Data Science · Data Engineering · TensorFlow · Keras · Forecasting · Python · Jira · Deep Learning · Machine Learning · Recurrent Neural Networks (RNN) · Convolutional Neural Networks (CNN) · Data Visualization · Text Mining · CI/CD · Scikit-Learn · Apache Airflow · Git · Natural Language Processing (NLP) · PyTorch · REST APIs · BigQuery · Google Cloud Platform (GCP) · MLOps · Time Series Forecasting · Docker · AI Agents

I worked as a Cloud Data Engineer on Proofs of Concept (PoCs) on Google Cloud Platform (GCP). My role involved leveraging advanced cloud solutions to process and analyze large amounts of data in a scalable and automated manner, with a focus on system efficiency and performance.

- Execution of Proofs of Concept (PoCs) based on Google Cloud (GCP) solutions, implementing massive and scalable data processing using BigQuery, Cloud Dataflow, and Vertex AI.

- Participation in projects integrating advanced cloud technologies for processing and analyzing large volumes of data, with a focus on automation and scalability.

- Contribution to the development of the entity's skills portfolio, helping team members upskill on cloud tools.

Skills: Data Analysis · SQL · Python (Programming Language) · Machine Learning · MLOps · Google BigQuery · Google Cloud Platform (GCP) · Google Cloud Dataflow · Google Cloud