About Me

Senior Generative AI Engineer

appliedAI Initiative GmbH • Munich, Germany

Oct 2025 - Present

• Architected and led the development of a retail-specific sales intelligence platform for Conrad, built on a multi-agent engine using LangChain and LangGraph. Applied a Swarm/Handoff design pattern to orchestrate four specialised agents covering customer insights, lead generation, note capture, and product discovery, automating the full sales lifecycle end to end.
• Designed and led the development of an AI-powered supply chain automation system for Siemens Energy, leveraging a multi-agent architecture with a Supervisor design pattern for orchestration to autonomously process emails, extract structured RFQ requirements, and score and rank supplier offers, significantly reducing manual effort in procurement workflows.

Senior Machine Learning Engineer / Tech Lead

ProSiebenSat.1 Media SE • Munich, Germany

Nov 2024 - Sep 2025

• Built a Databricks data pipeline to compute audience segment shares, enabling optimised ad campaign placements across channels.
• Organised cross-departmental knowledge sharing sessions to accelerate knowledge transfer and strengthen technical synergy within the data organisation.

Machine Learning Engineer / Tech Lead

ProSiebenSat.1 Media SE • Munich, Germany

Apr 2022 - Oct 2024

• Led the development of a RAG-powered internal chatbot allowing sales teams to instantly retrieve company knowledge, saving up to 40 minutes per person per day.
• Improved content categorisation using IAB-standard text classification, enabling precise contextual targeting and cookieless advertising strategies.
• Trained user activity forecasting models and engineered a scalable ML pipeline orchestrated with Metaflow and tracked via MLflow.
• Developed a TV attribution algorithm to measure and quantify the impact of TV ad campaigns on client website performance.

Associate Data Scientist

ProSiebenSat.1 Media SE • Munich, Germany

Feb 2021 - Apr 2022

• Migrated an on-premises TV Attribution system to AWS, reducing infrastructure and maintenance costs by 60%.
• Developed and deployed a web application to monitor complex data pipelines, reducing daily operational overhead by 20%.
• Trained a SageMaker classification model to predict users' age and gender from viewing behaviour, improving ad targeting accuracy.

Master Thesis - Data Scientist (Grade: 1.0)

Porsche Digital GmbH • Ludwigsburg, Germany

Mar 2020 - Sep 2020

• Designed and trained an Improved Deep Embedded Clustering (IDEC) model for unsupervised anomaly detection on acoustic signals using Keras.

Working Student - Data Analyst

Siemens AG • Karlsruhe, Germany

Nov 2018 - Apr 2020

• Built real-time Qlik Sense dashboards to monitor 8D manufacturing nonconformance reports, improving management visibility and response times.

M.Sc. Computer Science

Karlsruhe Institute of Technology – KIT • Karlsruhe, Germany

Oct 2018 - Sep 2020 • Grade: 1.8

Focus: Machine Learning, Computer Vision and Deep Learning

• Trained an LSTM-based RNN for movement prediction and anomaly detection in shipping traffic.
• Applied data augmentation and transfer learning to train a CNN for face recognition with a small dataset.

B.Sc. Computer Science

Karlsruhe Institute of Technology – KIT • Karlsruhe, Germany

Oct 2014 - Sep 2018 • Grade: 2.4

Focus: Software Development

• Implemented a web-based course enrollment system using Play and Java.

Diploma in Mathematics and Physics

Tunis Preparatory Engineering Institute – IPEIT • Tunis, Tunisia

Sep 2011 - Jul 2013 • Grade:

Focus: Mathematics and Physics