To have the most relevant environmental scores, we need a rich database, relying of multiple sources (relying on a single source is risky), a efficient impact computation (mostly the 2° aligment and the biodiversity impact) and frequent deep analyses on the quality of our scoring.

Among the ~30 companies employees, the data science team (4 people) is part of the IT team (10 people) and is looking for an apprentice to:

• Improve our data pipelines (we fetch financial and ecological data from different sources). The cleanest the data we fetch, the better is our final coverage of companies worldwide.

• Search and collect new data sources (GHG emissions from companies, assets, tons of products produced, consumed, companies facilities, etc.

• Leverage on Large Language Models (ChatGPT like) API and Langchain framework to improve our semi-structured data parser (like ESG reports)

• Automatically detect discrepancies between the default model and all refinements brought by our analysts.

• Benchmark our scores against academic environmental papers and competitors

Minimum: Data Science, data analysis, data engineering, Python, Pandas, SQL, Git, interest in environmental issues.

Optional: Machine Learning, Docker, Langchain, FastAPI, ElasticSearch, Streamlit, Spark, Kubertnetes, Celery, , Ruby, Financial or ecological knowledge.

20 min call, then 1H30 technical interview.