STAGE Data Scientist - Machine learning applied to solar PV

Date de mise à jour de l’offre


FEEDGY (previously QUANTOM) is a Paris-based PME specialised in the analysis and optimisation of performances of solar photovoltaic (PV) installations. In the last 5 years, the FEEDGY R&D experts have been working on developing advanced algorithms for analysing the performance of PV installations and particularly on detecting anomalies and diagnosing their root causes. Solutions developed by FEEDGY are data-driven and rely on machine learning to create insights that are impossible with traditional methods used in the PV sector. FEEDGY Analytics is a newly created division to reflect our ambitions to become a major actor in the analytics market. FEEDGY Analytics is launching an ambitious development project to open our innovations to the market.

Description de la mission

Solar PV plants convert solar energy to electricity. While operating, a large amount of data is created in the form of time series. Example variables include electrical variables (power, current, voltage…) and environmental variables (solar radiation, wind speed, air temperature…). The electrical variables describing the behaviour of plants is directly influenced by the environmental variables. Various phenomena such as defects or faults impact the behaviour of plants provoking losses in production that reduce the profitability of the PV plants.

During the apprenticeship, the apprentice will contribute to the development and maintenance of regression and classification ML models. Example of tasks include:
• Collaborating with ML engineers in the development of high complexity ML models (several areas of expertise involved, deep neural network models)
• Conduct research that can lead to the development of ML prototypes and proof of concepts
• Documentation of new ML models
• Assess the effectiveness of data sources and data-gathering techniques and improve data collection methods
• Mine, analyse and label data from company databases to drive optimization and improvement of product development
• Collaborating with software engineers to translate ML models into production systems.
• Collaborating with product managers and other stakeholders to identify new areas of potential impact.
• Querying, analysing data, and setting up monitoring and data quality assurance (including putting in place data cleaning methods if necessary)
• Maintenance and evolution of existing ML models

Profil recherché

• Masters'student in the field of data science
• Basic knowledge of a variety of machine learning techniques (clustering, decision tree learning, artificial neural networks, etc.) and their real-world advantages/drawbacks.
• Basic knowledge of advanced statistical techniques and concepts (regression, properties of distributions, statistical tests and proper usage, etc.) and experience with applications.
• Experience in building and maintaining data pipelines and collecting data from multiple sources and formats, and working with data science-based services that provide an API
• Experience using Pytorch or Tensorflow
• Good knowledge of data science libraries (Pandas, Numpy, Matplotlib etc.)
• Good analytical and quantitative skills with experience of SQL and python.
• Good presentation skills
• Communication skills in English
• Sufficient oral French to participate collective activities / willingness to learn

Niveau de qualification requis

Bac + 4/5 et +
  • Employeur
  • Secteur d’activité de la structure
    Emploi - Economie - Innovation - Numérique
  • Effectif de la structure
    De 21 à 50 salariés
  • Site internet de la structure
  • Type de stage ou contrat
    Contrat d'apprentissage
  • Date prévisionnelle de démarrage
  • Durée du stage ou contrat
    Supérieur à 6 mois
  • Le stage est-il rémunéré ?
  • Niveau de qualification requis

    Bac + 4/5 et +
  • Lieu du stage
    83 Rue la Fayette
  • Accès et transports
    Proximity to public transportation