Data Scientist

About the Role: 

Part of the Allianz Group, Allianz Trade is the global leader in trade credit insurance. For more than 10 years, the Group Data Analytics department has been conducting advanced data analysis and building data science and machine learning models that power the company's core processes: risk assessment, pricing, underwriting, claims management and collection.

As an insurance company dedicated to global trade, Allianz Trade is the perfect playground for data science. Evaluating businesses around the world puts us at the crossroads of countless information sources, and therefore of data, models and opportunities. Today, our scope spans the full spectrum of modern AI, from classical machine learning to generative AI and agentic systems, with a strong focus on industrialization, monitoring and responsible AI in a regulated European environment (EU AI Act).

As a data science intern, you will join a global team of 30+ data scientists, machine learning engineers and modelling experts, and contribute to our R&D, development and monitoring roadmap, working on real use cases with measurable business impact.
 

Key Responsibilities: 

• Contribute to the modelling, pricing, GenAI and monitoring roadmap of the department. Depending on your profile and interests, you will work on one or more assignments.
• In actuarial science and technical pricing, you may contribute to the development and refinement of our pricing models: estimating risk premiums through frequency and severity modelling, improving loss ratio and claims cost predictions, and exploring how machine learning techniques (GLM/GAM extensions, gradient boosting, calibration methods) can complement traditional actuarial approaches in trade credit insurance.
• In agentic AI development, you may design, prototype and evaluate LLM-based agents and multi-agent workflows (tool use, RAG, orchestration) for business use cases such as underwriting support, bond management or document processing.
• In GenAI monitoring and evaluation, you may build state-of-the-art tooling to track generative AI use cases in production: usage analytics, output quality evaluation, drift and hallucination detection, cost and performance dashboards.
• In ML model monitoring, you may strengthen the monitoring of our machine learning and pricing model portfolio by automating performance reviews, data-drift analyses and reporting pipelines.
• In applied R&D, you may benchmark emerging techniques (new foundation models, fine-tuning, evaluation frameworks, agent protocols) and assess their relevance for Allianz Trade, from proof of concept to production readiness.

Profile/Required Skills: 

• Data Science skills:

- Python (good knowledge is a plus, willingness to learn more is mandatory) - Data science & machine learning algorithms and best practices - Standard Python data science libraries (Pandas, scikit-learn, Matplotlib) - Exposure to generative AI / LLM frameworks (RAG, agents, evaluation) is a plus - Basics of Git and SQL are a plus - Background in statistics or actuarial science (GLM, frequency/severity modelling) is a plus for pricing assignments