Victor Gerard

Context:

Within an international context, I was responsible for the performance of FX+ Payments, a critical platform for automated foreign exchange in international payments.

Actions:

- Cash Management Officer Deck: Develop a Python tool to generate tracking for hundreds of client portfolios based on defined metrics, across multiple countries and regions.

- TFX Data Pulse: Development of an automated Python tool that informs all stakeholders in real-time of Tableau production data via email every morning.

- API Architecture Setup: Set up a Python client "tableauserverclient" within JupyterHub to provide real-time access to Tableau data, eliminating the need for manual ETL. Allows direct data processing.

- Development of advanced interactive Tableau dashboards.

- General improvement of processes implemented with Python.

Results:

Selected 3 months after my arrival for a competition of the most impactful initiatives of the year. The tools were used in production and praised by my superiors/colleagues.

Context:

I joined Thales AMS LAS in Rungis to assist a team responsible for a middleware for civil aviation. We worked using the Agile methodology.

Actions:

- Python tool for mass data cleaning according to team directives on a shared NAS drive.

- Indexing of unstructured documents in Elasticsearch with the creation of visualizations on Kibana. Use of the ELK stack in a Python environment.

- Python tool for automated tracking of code coverage on git, allowing easy comparison of project versions.

- Contribution to strengthening company culture by initiating and participating in collaborative projects.

Results:

Tools used internally.

Context:

I joined the Data department of Crédit Agricole Titres to assist the business units of the subsidiary.

Actions:

- Sentiment analysis based on the completeness of comments (100K+) from 6 applications of the BNP Paribas and Crédit Agricole group. Use of scraping techniques, Natural Language Processing (NLP), and creation of an advanced Power BI dashboard. All this was done in Python, using the Transformers architecture (BERT) for sentiment analysis.

- Automation of data reconciliation with Power BI and Power Query.

Results:

Full confidence from project stakeholders.