Working Student – GenAI / LLM Evaluation – Agentic AI / NLP (f/m/d)

Position Description

In this role, you will:

• Support evaluation of agentic AI systems and LLM-based NLP features, including qualitative and quantitative analysis.

• Create, curate, and maintain datasets for benchmarking, regression testing, and scenario coverage.

• Extend and improve internal evaluation frameworks (metrics, dashboards, automated test runs).

• Contribute to end-to-end testing of GenAI features within the in-car experience, including integration and validation workflows.

• Document findings, track model/system changes, and communicate results clearly to the team.

• Collaborate with engineers and researchers to translate evaluation insights into actionable improvements.

What you will need to succeed:

• Ongoing Bachelor’s or Master’s studies in Computer Science, AI/ML, Data Science, Computational Linguistics, or a related field.

• Hands-on programming skills in Python and a solid understanding of basic ML/NLP concepts.

• Interest in GenAI / LLMs, agentic systems, and evaluation of non-deterministic AI behavior.

• Experience with data handling and dataset creation (labeling, preprocessing, quality checks).

• Familiarity with software testing concepts (e.g., unit/e2e testing, CI) is a plus.

• Good written and spoken English communication skills.

• The successful candidate will be based in Karlsruhe, Germany.