In this position, you will support the development and evaluation of AI-driven approaches to enhance Pointsharp’s IT security products, with a focus on automated classification of sensitive data. You will work on the application of Large Language Models (LLMs) to analyze and classify file content and email communication into different security levels.
The role offers a hands-on opportunity to combine theoretical knowledge with practical implementation in an industry-relevant use case, contributing directly to the evolution of intelligent security solutions.
Student Position Research/AI
DE- Karlsruhe Office, DE- Freiburg i. Br. Office
Part-time
Working Student
About the role
What you'll do
- Investigate how Large Language Models (LLMs) can be used to classify files, messages, and emails into security levels
- Design and prototype approaches for automated classification of confidential data
- Evaluate performance, quality, and robustness of LLM-based classification methods
- Analyze challenges related to context understanding, accuracy, and security
- Document findings and derive recommendations for integrating AI-based classification into existing products
What we're looking for
- You are studying computer science, business informatics, IT security, data science, or a related field
- You have a strong interest in AI, especially in Large Language Models and their practical applications
- You have analytical thinking skills and enjoy working on complex, real-world problems
- Ideally, you have some experience with Python, NLP, or machine learning frameworks
- You work independently, structured, and are able to document your results clearly
- You are fluent in German and English
- You are available as a working student or for a thesis project (Bachelor’s or Master’s)
- A highly relevant, future-oriented topic at the intersection of AI and cybersecurity
- The opportunity to contribute to real product innovation
- Flexible working conditions and support from experienced professionals
- A collaborative environment in our Karlsruhe or Freiburg office (with partial remote work possible)