Applied Research Scientist in Robot Learning & Manipulation

What are we looking for? 

We do not expect you to match every keyword. We are looking for depth, curiosity, and the ability to turn research ideas into working robotic systems. 

Essential 

• PhD in Robotics, Artificial Intelligence, Machine Learning, Computer Vision, Control, or a related field; or equivalent research experience.  

• Strong background in machine learning or deep learning applied to robotics.  

• Experience with robot learning, manipulation, embodied AI, learning for control, or AI-based robotic perception/action.  

• Strong programming skills in Python.  

• Ability to work with real robotic systems, not only offline datasets or simulation.  

• Scientific and technical maturity: ability to define problems, test hypotheses, analyse results, and communicate findings.  

• Good written and spoken English.  

Highly valuable 

• Experience with imitation learning, behaviour cloning, reinforcement learning, offline RL, diffusion policies, transformers, VLA models, or foundation models for robotics.  

• Experience with ROS/ROS 2.  

• Experience with PyTorch, TensorFlow, or similar frameworks.  

• Experience with robotic arms, mobile manipulators, force/contact-rich manipulation, bimanual manipulation, or human-robot collaboration.  

• Publications, open-source contributions, project demos, or research prototypes in robot learning or embodied AI.  

• Experience in European or collaborative R&D projects.  

This role is probably a good fit if… 

• You have worked on AI methods that make robots perceive, decide, manipulate, adapt, or learn.  

• You enjoy both research and implementation.  

• You are comfortable moving between papers, code, experiments, robots, and project discussions.  

• You want to work in applied research: scientifically ambitious, but connected to real systems and industrial impact.  

• You are attracted by the opportunity to contribute to a European flagship effort in robotics foundation models.  

This role is probably not the best fit if your main experience is limited to… 

• Generic object detection without a connection to robot action or manipulation.  

• LLM chatbot development without robotics.  

• Pure navigation, PLC programming, or classical control with no AI/robot-learning component.  

• Simulation-only research with no interest in deploying on real robots.  

• Software front-end or general AI engineering unrelated to physical robotic systems.  

What Eurecat offers 

Eurecat is a leading technology centre in Catalonia, working at the intersection of research, technology transfer, and industrial innovation. 

In our Robotics and Automation Unit, we develop applied research in service and industrial robotics, including manipulation, human-robot collaboration, perception, cognition, robot behaviour, and autonomous systems. 

You will join a team with strong expertise in European projects, industrial collaborations, and real-world robotic demonstrators. Our laboratory is based in Cerdanyola del Vallès, just outside Barcelona, at the heart of a vibrant and growing robotics and AI ecosystem that brings together leading research centers, universities, technology companies, and innovation initiatives.

Additionally, you will have these conditions and benefits: 

• Permanent contract.  

• Hybrid work.  

• Flexible schedule.  

• Shorter workday on Fridays and summer schedule.  

• Flexible compensation package: health insurance, transport, lunch vouchers, training, kindergarten, and other benefits.  

• Access to Eurecat Academy courses.  

• Language training in English, Catalan, and Spanish.  

• Participation in a major European robotics and AI project with leading academic and industrial partners.  

• The opportunity to build long-term research lines beyond a single project.