Agentic AI Research Engineer (m/w/d)

Heusenstamm, Germany

KION Group

We are a leading supplier of forklifts and warehouse equipment as well as automation technology and software solutions for the optimization of supply chains.

View all jobs at KION Group

Apply now Apply later

Our Basics:

A challenging and dynamic work environment within a global leader in automated solutions

Opportunities for continuous learning and professional development

Competitive salary and benefits package

Supportive and collaborative team culture

What we offer:
  • Development of autonomous, robotic systems for intralogistics using artificial intelligence models and methods.
  • Construction of mobile robots as demonstrators and prototypes in an existing laboratory environment.
  • Integration of generative AI methods (e.g., LLMs) into the control of mobile robots.
  • Investigation of the applicability of Agentic AI in intralogistics.
  • Development of software architectures for linking AI models with robotic systems.
  • Transfer of innovations from research to product development.
  • Management and administration of innovation projects together with internal and external partners.

Tasks and Qualifications:
  • Master's degree in Computer Science or Engineering, ideally with a PhD
  • Experience in the development of mobile robotic systems using ROS2, and their simulation in NVIDIA Omniverse
  • Programming experience with Python and C++
  • Knowledge in the field of Artificial Intelligence, particularly in the handling of Generative AI methods (e.g., LLMs)
  • Experience with Agentic AI is advantageous
  • Several years of experience in leading and managing innovation projects
  • Excellent English presentation skills
Apply now Apply later

* Salary range is an estimate based on our AI, ML, Data Science Salary Index πŸ’°

Job stats:  2  0  0

Tags: Architecture Computer Science Engineering Generative AI LLMs PhD Python Research

Perks/benefits: Career development Competitive pay

Region: Europe
Country: Germany

More jobs like this