About me

My name is Tanguy Esteoule, I currently work as a Data Scientist in the company meteo*swift, specialized in artificial intelligence in the field of energy transition. My work focuses specifically on wind power forecasting. In particular, I study the consideration of interdependencies and interactions between wind turbines for forecasting.

I did my Ph.D., Adaptive Multi-Agent Systems for Wind Power Forecasting, in the SMAC team at IRIT, computer science laboratory in Toulouse, France.

I am naturally curious and I have always been passionate about music.

Experiences

Data Scientist

2016 - Present
meteo*swift, Grenoble

Initially Ph.D. thesis in industrial relations (CIFRE Agreements) then full-time job since 2020

  • Design and development of new wind power forecasting models
  • R&D on new products related to the company’s field of activity
  • Integration of the research work carried out during the thesis into the wind farms managed by the company
  • Development of internal tools
  • Writing internal scientific reports

Computer Science Teacher

2017 - 2019
University of Toulouse

Python practical work: Introduction to the basic structures of algorithmic, implemented in Python (28h)

Data Scientist Intern

2016
Airbus, Toulouse

Data-Mining technics for analysis of Aircraft logs from On-board In

  • Analysis of aircraft logs produced by the FSA-NG, an embedded system for flight, maintenance and ground operations
  • Search for correlations between events that occurred during the flight by Association Rule learning
  • Development of a pattern detection tool, implemented in Java

Education

Ph.D. in Artificial intelligence

2016 - 2019
University of Toulouse

Thesis: Adaptive Multi-Agent Systems for Wind Power Forecasting

  • Training a wind power forecasting model from erroneous numerical weather predictions
  • Design and development of an Adaptive Multi-Agent System (AMAS) in charge of wind power forecasting
  • Detection and consideration of wake effect constraints between wind turbines
  • Works on weather ensemble forecasting in partnership with the French Weather Research Centre (CNRM)

Jury:

  • Pierre PINSON, Professor, Technical University of Denmark (Reviewer)
  • Yves DEMAZEAU, Director of Research, Université Grenoble Alpes (Reviewer)
  • Gauthier PICARD, Professor, Mines de Saint-Etienne (Examiner)

MSc in Computer Science

2013 - 2016
Ensimag, Grenoble

Specialization in Embedded Systems

Graduate Exchange

2015
Pontifical Catholic University of Peru

Exchange semester in a Peruvian university

  • Data science and programming course

Projects

Movizz - Online multiplayer game based on film dialogues (Django)
Wikipedia Meter - A twitter bot that publishes every day a word cloud of the most read articles on Wikipedia France
This album cover does not exist - Generating realistic album covers with artificial intelligence (with Generative Adversarial Networks)
Kajika - Web-based multi-agent system visualization application (Developed by Inès Louahadj under my supervision)

Publications

Adaptive Multi-Agent Systems for Wind Power Forecasting
Tanguy Esteoule
Ph.D. thesis, Université Paul Sabatier (2019)
Improving Wind Power Forecasting through Cooperation: A Case-Study on Operating Farms
Tanguy Esteoule, Carole Bernon, Marie-Pierre Gleizes and Morgane Barthod
International Conference on Autonomous Agents and MultiAgent Systems (AAMAS) (2019)
A Cooperative Multi-Agent System for Wind Power Forecasting
Tanguy Esteoule, Alexandre Perles, Carole Bernon, Marie-Pierre Gleize and Morgane Barthod
International Conference on Practical Applications of Agents and Multi-Agent Systems (pp. 152-163). Springer, Cham. (2018)

Skills & Proficiency

Python

LaTeX

Java / C#

HTML5 / CSS3 / JS

C / C++

Photoshop


Frameworks & Libraries

Data science

Scikit-learn / Keras / TensorFlow

Web

Vue.js / React.js / node.js / Dango

Database

MySQL / MongoDB