Feasibility Study for Code Generation
Agentic AI transforms code generation by acting as a proactive collaborator: planning, decomposing tasks, and refining outputs with human feedback. This study explores its feasibility, benefits, and integration in real workflows.
Factsheet
- Schools involved School of Engineering and Computer Science
- Institute(s) Institute for Data Applications and Security (IDAS)
- Research unit(s) IDAS / Applied Machine Intelligence
- Funding organisation Others
- Duration (planned) 01.03.2025 - 31.08.2025
- Head of project Prof. Dr. Souhir Ben Souissi
- Partner Golem Technologies Europe (Switzerland) SA
Situation
The Generative AI Lab at BFH will conduct a comprehensive concept and feasibility study to address critical questions in the application of NLP models for regulated sectors. The study will explore how multilingual models can be fine-tuned for accurate SQL, API, and code generation while ensuring compliance with Swiss and EU data privacy standards through on-premises deployment. Additionally, it will investigate measures to reduce hallucinations and enhance the reliability of AI outputs, as well as design scalable solutions for integration with existing ERP databases such as SAP, Oracle, and IBM. A particular focus will be placed on the role of agentic AI, which extends beyond static code generation. This paradigm will be assessed for its potential to deliver more robust, trustworthy, and adaptive AI-driven coding workflows in highly regulated environments.
