Nesrine Haddad

• • Lead and coordinate multiple AI initiatives from ideation to deployment, ensuring alignment between technical feasibility and business needs.

• • Design, prototype, and deploy end-to-end AI systems leveraging AWS cloud services for intelligent document processing, RAG architectures, and LLM based automation.

• • Build and orchestrate data pipelines and machine learning workflows leveraging Dataiku for preprocessing, feature engineering, and model experimentation with downstream AWS services.

• • Conduct FinOps monitoring for model usage, token costs, and performance metrics.

• • Drive AI awareness and adoption programs, organize workshops, and support training for procurement stakeholders.

• • Act as a bridge between technical and business teams, ensuring that AI models translate into operational value.

During my internship, I focused on optimizing energy consumption for ovens in the paint shop. This project involved several key tasks:

• • Optimize Heating Cycles: Develop machine learning models to predict the time required for each sensor to reach target temperatures, enabling precise identification of optimal ignition timings.

• • Forecast Temperature Behavior During Cooling Phases: Build forecasting models to predict temperature evolution during cooling phases, allowing operators to accurately determine the temperature at any given time.

• • Develop an Iterative Predictive System: Combine temperature and time prediction models into an iterative algorithm, providing operators with forecasts at multiple timestamps spaced by 10-minute intervals. This predictive system is integrated into a web application to create a user-friendly tool for improved planning of oven operations.

• • Real-Time Monitoring: Create dashboards for visual insights into historical temperature trends, phase durations, and correlations between temperature changes and environmental conditions.

• • Enhance Quality Control: Design a dashboard to visualize defect locations across vehicle parts, improving the overall quality control process.

• • Carbon demand update on long-term horizon.

• • Improvement of carbon forecast accuracy and digitalization of company KPIs through implementation of Power BI reports.

• • Preparation of the carbon executive demand review in the S&OP monthly loop.

• • Digitalization of company forecasting tools.

Specifically employed in the Data Management Department of Military Supply Chain Division, i was in charge of:

• • Treatment and cleaning of the available datasets.

• • Measurement of supply forecasts uncertainty.

• • Development of KPIs to assess the reliability of the several suppliers and improvement of the company's "On Time Delivery" indicator regarding their customers.

• • Development of traceability and update algorithms.

• • Development of a decision support tool to assess the reliability of the supply forecasts through the implementation of Power Bi reports.