Summary
KTH Royal Institute of Technology invites applications for a Postdoctoral Researcher in Robot Learning for Manipulation. This position focuses on exploring state-of-the-art Vision–Language–Action (VLA) models for various manipulation tasks, emphasizing whole-body motion and/or dexterous, multi-fingered hands to enhance perception and control. The role involves solving complex, contact-rich tasks requiring coordinated robot behaviors in dynamic environments. This is a research-focused, time-limited appointment intended as a first career step after a dissertation.
Designation
Postdoctoral Researcher
Research Area
- Robot Learning for Manipulation
- Vision–Language–Action (VLA) models
- Deep Learning
- Reinforcement Learning
- Imitation Learning
- Robot Perception
- Navigation
- Grasping
- Dexterous or Mobile Manipulation
- Whole-body control
- Humanoid or Multi-DOF platforms
- Multi-fingered hands
- Tactile sensing
Location
Stockholm, Sweden
Eligibility/Qualification
Required Qualifications
- A doctoral degree or an equivalent foreign degree, to be met no later than the time of employment decision.
- Research expertise in machine learning/robotics.
- Documented strong programming skills.
- Outstanding research and publication list.
- Well-developed analytical and problem-solving skills.
- Very good knowledge of the English language (written and oral).
Preferred Qualifications
- A doctoral degree or an equivalent foreign degree obtained within the last three years prior to the application deadline.
- Good pedagogical ability.
- Awareness of diversity and equal opportunity issues, with specific focus on gender equality.
- Independent work ethic combined with strong collaborative skills.
Job Description
The postdoctoral researcher will engage in cutting-edge research on robot learning for manipulation. The primary focus is on leveraging Vision–Language–Action (VLA) models to enable robots to perform complex tasks, including tabletop, mobile, or dexterous manipulation. A key aspect of the research involves utilizing whole-body motion and/or multi-fingered hands to improve perception and control, allowing robots to tackle contact-rich tasks that demand coordinated base, arm, and hand behaviors in dynamic settings. The role requires a strong background in robotics and machine learning, with practical experience in areas such as deep learning, reinforcement learning, imitation learning, robot perception, navigation, grasping, and various forms of manipulation. Experience with whole-body control, humanoid platforms, multi-fingered hands, tactile sensing, or learning-based manipulation in both simulation and real systems is highly valued.
How to Apply
To apply for this position, candidates must log into KTH’s recruitment system. Ensure your application is complete according to the instructions provided in the advertisement. The application must include:
- CV: Including relevant professional experience and knowledge.
- Copies of Diplomas and Grades: From previous university studies. Translations into English or Swedish are required if original documents are not in these languages.
- Brief Account: A statement (maximally two pages long) explaining your motivation for conducting research, your academic interests, and how they relate to your previous studies and future goals.
Your complete application must be received by KTH no later than the last day of application, midnight CET/CEST (Central European Time/Central European Summer Time).
Last Date for Apply
01.Nov.2026
Apply Link
Apply for Postdoc in Robot Learning for Manipulation
