Nathan Tessot

I led an end-to-end data analysis project focused on customer reviews collected via the Google Maps Reviews API. I retrieved and processed large volumes of reviews, then applied statistical methods to highlight actionable trends: recurring complaints, satisfaction drivers, and evolutions based on location and time. The objective was to transform scattered and unstructured customer feedback into clear, evidence-based insights, directly actionable by local businesses. I translated these results into concrete recommendations to help merchants improve their customer experience, prioritize the most important issues, and refine their overall strategy. This project combines data collection via API, statistical analysis, and the ability to present technical results in clear, business-oriented language.

Design and development of a modern and interactive website explaining RAG (Retrieval-Augmented Generation) to EDF's R&D teams. The challenge: making a complex AI concept truly understandable for a technical but non-specialist audience. I structured the content as a visual and progressive journey rather than a dense block of text, relying on interactivity and a clean design to maintain researchers' attention while faithfully representing how RAG works. The result: an educational experience that combines scientific rigor, clarity, and visual quality. This project demonstrates my ability to grasp technical subjects, design intuitive user experiences, and develop polished, responsive front-end interfaces that effectively convey an idea.

Development of a web application that automates the creation of visuals for professional seminars using AI. The team was spending considerable time manually producing these visuals; therefore, I built a tool that streamlines this repetitive process from end to end — automatically generating custom visuals and freeing up the team for higher-value tasks. The project involved integrating AI generation into a functional and accessible web interface, as well as designing a workflow that can be easily used by non-technical individuals. It demonstrates my mastery of AI integration, process automation, and the creation of practical tools that solve a real, measurable time problem, rather than adding features for the sake of it.