Kurz gesagt
Du entwirfst und baust KI-Agenten, die regulatorische, rechtliche und technische Normdokumente in eine strukturierte Compliance-Wissensbasis überführen. Du strukturierst eine zentrale Compliance-Bibliothek in SharePoint, entwickelst Extraktionsagenten und baust Verifikations- und Validierungsschleifen auf. Außerdem testest und optimierst du die Agenten, verpackst sie zur Wiederverwendung und bringst Ideen für unsere KI-Workflows ein.
- Entwicklung von KI-Agenten auf Basis von LLMs
- Prompt Engineering
- Claude / Anthropic
- Token- und Context-Management
- Aufbau effizienter und zuverlässiger LLM-Workflows
WORKING STUDENT (f/m/d) AI AGENT DEVELOPMENT — COMPLIANCE MANAGEMENT (KNOWLEDGE MANAGEMENT)
DACMA GmbH, Hermann-Blohm-Straße 3, 20457 Hamburg. Job ID: S-AI202607. Working student (Werkstudent), Hamburg headquarters.
We are seeking a highly motivated working student to help us build the next generation of our Compliance Management system. This is a hands-on AI engineering role at the intersection of applied AI and climate technology.
YOUR ROLE
You will design and build AI agents that turn regulatory, legal and engineering-standard documentation into a structured, trustworthy compliance knowledge base for our Direct Air Capture (DAC) projects. Working on Claude Enterprise and SharePoint, together with our project, engineering, finance and quality experts, you will:
- Design and build LLM-based AI agents that discover and retrieve specific data from official regulatory, legal and engineering-standard documents
- Structure a central compliance library in SharePoint, with clean metadata, versioning and full source traceability
- Develop extraction agents that read source documents and automatically compile a structured compliance register
- Build verification and validation loops so that every automated output is accurate, sourced, and audit-grade
- Iteratively test, refine and optimize the agents, including prompt design and efficient token and context management
- Package the finished agents for reuse and future deployment
- Contribute ideas to improve our AI workflows and engineering knowledge management
WHAT YOU BRING
Essential
- Enrolled in Computer Science, Data Science & Artificial Intelligence, Information Technology, or a related field (Bachelor's or Master's)
- Demonstrated hands-on experience building AI / LLM-based agents that you can show us (university, personal, or thesis projects)
- Solid hands-on knowledge of prompt engineering and working with large language models (Claude / Anthropic)
- Practical command of token and context management and of building efficient, reliable LLM workflows
- Precise, structured and methodical way of working; output must be accurate and audit-grade
- Good communication skills in English
Nice to have
- Experience with SharePoint / Microsoft 365
- Experience in structuring documents and data
- Interest in regulatory, standards or compliance topics
No legal or compliance background is required.
Location
Location: headquarters in Hamburg. We offer a collaborative startup environment, the chance to design and build a system from the ground up, and hands-on experience applying state-of-the-art AI to a real, high-impact engineering challenge.
Application by email to apply@dacma.de with Job ID S-AI202607.


