Arij Bouzid

Data Collection & Preparation: Gathered and preprocessed images of damaged vehicles along with related metadata (vehicle type, repair costs, etc.). AI Model Development: Built computer vision models to detect vehicles and localize damaged areas. Classified damage types and assessed severity. Predicted claim costs using regression algorithms. Model Evaluation: Validated model accuracy using real-world or simulated datasets. Prototyping: Designed and developed a prototype interface or API for integration into insurance company systems.

– Developed InsightAI, a predictive AI system that evaluates the impact of commits and predicts regression risks. – Implemented machine learning algorithms and a CI/CD pipeline to automate change evaluation, thereby optimizing the development workflow.

– Implemented the ETL process using Talend for extracting, transforming, and loading data into PostgreSQL. – Prepared data by applying the five phases of the CRISP-DM methodology. – Developed interactive dashboards with Power BI for dynamic data visualization. – Contributed to the design and development of a web application to clearly display relevant data and predictions generated by the machine learning system.