Company Description
Do you want beneficial technologies being shaped by your ideas? Whether in the areas of mobility solutions, consumer goods, industrial technology or energy and building technology - with us, you will have the chance to improve quality of life all across the globe. Welcome to Bosch.
Job Description
We are seeking an applied Research Scientist to join our team.
In this role, you will lead the design and development of large language models (LLMs)-based intelligent agent systems tailored for Bosch's industrial manufacturing scenarios.
You will leverage LLMs, agent development frameworks, and advanced NLP techniques to enable high-precision task automation, intelligent reasoning, and decision support.
You will architect and implement end-to-end agent capabilities—including function calling and tool use, memory mechanisms, task planning, and retrieval-augmented generation—to combine general AI capabilities with Bosch's deep industrial knowledge.
A key part of this role is driving post-training and fine-tuning efforts (e.g., instruction tuning, reward modeling, domain adaptation) that transform foundation models into high-accuracy, production-ready agents for real-world factory use cases.
Beyond solution development, you will continuously scout and evaluate the latest advancements in LLMs, agent frameworks, and knowledge-enhanced AI to ensure our solutions remain state-of-the-art.
By integrating cutting-edge research with Bosch's domain expertise, you will help close the "last mile" of AI deployment in industrial environments—delivering robust, reliable, and high-precision agent performance in production-level applications.
Qualifications
Design, develop, and optimize LLM-powered agent systems—including memory, planning, and reasoning capabilities—for high-precision task automation and decision support in industrial manufacturing scenarios.- Execute post-training and alignment of LLMs—including supervised fine-tuning (instruction tuning, domain/task adaptation), parameter-efficient tuning methods (e.g., LoRA/QLoRA), reward modeling, and preference optimization—to deliver controllable, high-accuracy behavior in industrial applications.
- Advance RAG systems beyond baseline implementations by optimizing retrieval, chunking, reranking, and grounding to achieve high-accuracy, domain-adapted performance in industrial scenarios.
- Collaborate with domain experts to transform industrial data into actionable insights using advanced NLP techniques
- Validate and benchmark agent performance in production-like environments, ensuring robustness and efficiency
- Contribute to scientific publications, patents, or technical whitepapers as part of Bosch's innovation initiatives
Basic Qualifications
- PhD or Master's degree in Computer Science, Artificial Intelligence, NLP, or related fields, with 3+ years of working experience in building real-world NLP or agent systems
- Proficient in Python and widely used LLM toolchains—covering model development (Transformers, PyTorch) and agent orchestration frameworks (LangChain, LangGraph, or equivalent).
- Solid understanding of LLM post-training and alignment workflows, including supervised fine-tuning (e.g., instruction tuning, domain adaptation) and reward/preference model optimization, with hands-on experience in at least part of this pipeline.
- Solid understanding of agent architectures (e.g., RAG, memory, tool use, planning) and their application in high-precision environments
- Proficiency in English for technical communication and collaboration in interdisciplinary teams
- Understanding of the deployment challenges of LLM/AI systems in production environments
________________________________________
Preferred Qualifications
- Experience applying LLM agents to industrial manufacturing domains
- Familiarity with knowledge graph related technologies- a plus
- Demonstrated contributions to top-tier AI/ML/NLP research (e.g., ACL, NeurIPS, ICML, ICLR)
- Some exposure to CV or multimodal models is a plus
