PhD – Agentic AI and Multi-Agent Systems

At Bosch, we shape the future by inventing high-quality technologies and services that spark enthusiasm and enrich people’s lives. Our promise to our associates is rock-solid: we grow together, we enjoy our work, and we inspire each other. Join in and feel the difference.

The Robert Bosch GmbH is looking forward to your application!

Responsibilities

As a PhD candidate with us, you will dive deep into the world of agentic AI systems and contribute significantly to the development of intelligent solutions for real-world challenges, combining cutting-edge research with direct industrial impact.; You will work on developing the next generation of AI systems for industrial applications and apply your expertise in practical projects.; Adapting foundation models efficiently and orchestrating multi-agent systems will be at the core of your work.; You will integrate knowledge graphs and retrieval-augmented generation (RAG) into innovative AI architectures.; A key part of your role involves researching and applying reinforcement learning–based training approaches for intelligent agents.; Last but not least, you will continuously validate your research results using real engineering use cases at Bosch, contributing directly to the advancement of modern enterprise processes.

Requirements

Education: excellent Master’s degree in Computer Science, Artificial Intelligence, Mathematics, or a comparable field; Experience and Knowledge:; strong knowledge of large language models (LLMs), foundation models, and deep learning, combined with hands-on experience in fine-tuning (e.g. SFT, RLHF/RLAIF) or parameter-efficient adaptation methods such as LoRA or adapters; experience with retrieval-augmented generation (RAG), including dense retrieval, reranking, and advanced architectures, as well as the integration of knowledge graphs, ontologies, or other knowledge engineering approaches (ideally with SPARQL, Cypher, or KG embeddings); familiarity with multi-agent systems or agentic frameworks (e.g. LangGraph, AutoGen, CrewAI), including aspects of agent safety, controllability, and human-in-the-loop approaches; strong programming skills in Python and experience with machine learning frameworks, ideally complemented by knowledge of reinforcement learning (e.g. RL, RLHF, policy optimization); experience in evaluating agentic systems using relevant frameworks or benchmarks, ideally complemented by contributions to scientific publications; Personality and Working Practice: you combine analytical thinking with a self-driven, structured approach, delivering results and taking ownership of your tasks, in international, collaborative environments, you stand out through strong communication and presentation skills; Languages : fluent in English