Master's Thesis: AI Agents & Simulation Development

At Bosch Rexroth everything revolves around the movement and success of our customers. With our cross-linkable drive and control technologies as well as digital solutions, machines and systems work efficiently, safely and efficiently. We offer our customers components, system solutions and services for mobile and industrial applications as well as for factory automation. In all areas we appreciate the know-how and commitment of our employees and enable dynamic careers in an international environment. Around 32,000 people in over 80 countries - and a corporate purpose: “We move industries to make our planet a better place. ” Do you want to move? We look forward to your application at Bosch Rexroth AG

Type of employment: Limited Working time: Full time Place of work: Horb am Neckar The model-based simulation development is time-consuming and requires extensive domain knowledge. As part of this six-month master thesis at Bosch Rexroth, you will examine how this process can be accelerated using coordinated AI agencies. The focus is on agentic engineering, the development of controllable agent-based workflows and the use of local AI models. Your tasks include: Analysis of patterns, anti patterns, and typical obstacles in building agent-based workflows, in particular to avoid and detect hallucinations Development of a harness engineering approach based on LangChain and LangGraph for the structured control of processes, tool accesses and test steps Investigation of the transition from cloud-based to local AI models including pre-training and fine tuning Assessment of performance in terms of token consumption and computing performance as well as protection of the quality of results by systematic evaluations and defined test cases Investigation from UX-Sicht, how developed simulations can be intuitively converted into productive use with or without support by an AI agent Development of a practical, controllable and transferable approach to AI-based simulation development

Education: You will complete a master's degree in Computer Science, Artificial Intelligence / Data Science, Computational Engineering, Engineering Informatics or a comparable technical degree course. Programming and AI: You have very good programming knowledge in Python and well-founded practical experience with LangChain and LangGraph (State Management, Routing, Tool Integration). LLM know-how: You bring knowledge in prompt engineering patterns, structured outputs and methods for avoiding and recognizing hallucinations as well as in dealing with open source LLMs (e.g. Llama, Mistral), fine tuning (e.g. PEFT/LoRA) and model adaptation. Further knowledge: Ideally you have: basic knowledge of model-based simulation and engineering workflows (e.g. Matlab/Simulink, Modelica, FMU) first experience with systematic LLM evaluations and benchmarking of token efficiency and hardware resources (inference latency, VRAM) Interest in human AI interaction and interface usability for productive use Experience with Git, Docker and Inference Engines for local models (e.g. vLLM, Ollama)

Personality and Operation: You work: analytically and solution-oriented and systematically disassemble complex engineering problems in order to map these software-related problems structured and quality-conscious with high standards of robust software architectures, fault treatment and scientific validation (Harness Engineering) independent and work intensively in current AI developments

Languages: You have fluent German or very good English knowledge in word and writing. Start: from February / March 2027 or by agreement Duration: 5 - 6 months

Work-Life Balance: Flexible work in terms of time, location and work model. Health and Sports: Wide range of health and sports activities. Childcare: childcare services. Employee discounts: Advantages for employees. Free space for creative work: free space for creative work. Social counselling and care: Social counselling and brokerage services.

The future personnel department or the department will be happy to inform you about the individual service catalogue. .