PhD Student Position - Learning Lab

Fundação Champalimaud; Lisboa; On-site; EUR 20,000 - 35,000; Full time

Job summary

Fundação Champalimaud in Lisbon is offering a full-time Fellowship position aimed at PhD students within the Marie‑Curie Doctoral Network (ELEVATE). The role focuses on computational neuroscience and the development of neural controllers for robotic systems.

Candidates must have a Master's degree in a relevant field and possess strong coding skills in machine learning frameworks. The Fellowship includes opportunities to publish research and participate in international conferences, providing a stimulating work environment.

Qualifications

• Strong coding skills and experience with machine learning frameworks are essential.
• Must be able to think creatively and analytically across disciplines.
• Good English communication skills are required.

Responsibilities

• Develop neural network controllers for robotic systems.
• Explore neuromorphic implementations and contribute to research publications.
• Participate in training networks and international conferences.

Skills

Coding skills for programming neural networks

Machine learning frameworks (e.g., PyTorch, Jax)

Analytical thinking

Collaborative work

Communication skills in English

Education

Master's degree in physics, engineering, computer science, or related field

Job description

Location: Champalimaud Centre for The Unknown

Job Type: Full Time

Position: Fellowship

Application Starts: 16 Jun. 2026

As part of a Marie‑Curie Doctoral Network (ELEVATE), we are offering a PhD student position. The proposed project sits at the interface of computational neuroscience and neuromorphic engineering, and is embedded within a strong network of European academic and industrial partners.

Learned behaviors ultimately must be expressed through the control of complex, high‑dimensional, continuously time‑varying patterns of muscle activity. However, attempts to learn policies for continuous control of behavior in engineering settings, such as robotics, have often leveraged the massive scale of training experience that is possible in the context of parallelized, offline simulation of behavior in simulated physical environments to achieve success. The brain cannot afford to be so sample inefficient, and appears to have greatly simplified the problem of learning effective control policies by embedding sophisticated invariances at lower levels of sensorimotor control circuits in the brainstem and spinal cord, effectively placing knowledge (ie memory) about the relationship between current state of an effector system, outgoing control signals, and the future state of the system within the circuits that process motor commands themselves.

This project aims to interrogate the functional principles underlying such mechanisms, leveraging a combination of behavioral experiments in head‑restrained mice performing forelimb reaches, computational modelling, and neuromorphic implementation. We will instantiate the extracted control policies and their adaptation in neuromorphic algorithms and validate them on a robotic arm.

Tasks

• Study the relevant background literature from both engineering and neurorobotics fields. Develop low‑level, neural network, sensorimotor controllers of musculoskeletal plants. Assess whether such controllers possess known features of spinal circuits, and what advantages are conferred on robotic control by mimicking known functional and anatomical features of spinal systems.
• Explore neuromorphic implementations of developed controllers together with Prof. Sandamirskaya.
• Publish research articles and regularly participate in top international conferences to present your work.
• Complete an internship at ZHAW with Prof. Yulia Sandamirskaya.
• Participate in yearly retreats organized by the doctoral network participants.
• Support the dissemination of software tools and concepts.

The student will be part of the Learning Lab, led by Joe Paton, hosted at the Champalimaud Centre for the Unknown in Lisbon, Portugal. The Champalimaud Foundation is a private, non‑profit institute with a vibrant and highly international academic community. Many groups work in the field of systems and computational neuroscience, providing a stimulating and exciting work environment. The student will also interact with the training network ELEVATE and spend several months at Yulia Sandamirskaya’s lab at ZHAW, Switzerland.

STARTING DATE: Earliest starting date is January 1st, 2026.

Requirements

• Research Fields: Neurosciences, Computer Science, Physics, Engineering
• Education Level: Master Degree or equivalent

Skills / Qualifications

The candidate needs to have a Masters degree in physics, electrical/electronic engineering, computer science, mathematics, computational neuroscience or a related field.

Specific Requirements

• Strong coding skills for programming neural networks, machine learning and machine learning software frameworks (e.g. PyTorch or Jax) is a must.
• The ability for creative and analytical thinking across discipline boundaries and abstraction levels is a must.
• Experience with control theory and spiking neural networks is a plus.
• Ability for collaborative work, interdisciplinary and cross‑topical thinking.
• Very good communication skills in English, both spoken and written.

Language

The salary and benefits will follow standard EU rules. PhD students will receive their degree from the NOVA University of Lisbon.

Selection Process

We are looking for a candidate with a strong relevant background and curiosity about how the brain works. Interested candidates should submit a detailed CV, a cover letter explaining their interest in the position (1 page), a transcript for their BSc and MSc, and the contact information of 2 references.

We strive to build a diverse work environment and encourage applications from all qualified individuals irrespective of their gender, age, cultural background, and disability status. Only candidates shortlisted for this position will be contacted for an interview.