Reinforcement Learning Engineer

Our client is pioneering Level 4 certifiable autonomous driving solutions, tailored for public transport and designed with safety at the core. By leveraging cognitive intelligence and cutting-edge AI based on German research, we create autonomous systems that make logical, explainable decisions in complex road scenarios. Our mission is to enable sustainable, safe, and scalable mobility solutions, ensuring that autonomous technology can connect people everywhere—especially in rural areas and underserved communities.

As a Reinforcement Learning Engineer, you'll be instrumental in advancing our unique decision-making framework based on cognitive neuroscience. Your expertise in inference-driven AI, probabilistic modelling, and goal-directed behaviour will help us develop explainable, adaptive systems for autonomous driving.

Responsibilities:
• Design and implement decision-making architectures based on Active Inference, Bayesian models, and reinforcement learning principles.
• Develop generative models and inference-based systems to guide autonomous agents under uncertainty.
• Integrate concepts from cognitive robotics, predictive coding, and goal directed behaviour into scalable autonomous driving modules.
• Apply and extend the Free Energy Principle and planning-as-inference frameworks for real-world applications in perception and control.
• Model and simulate agent-based, hierarchical inference systems to support adaptive, real-time decision-making.
• Collaborate cross-functionally with neuroscience-inspired perception, planning, and systems teams to ensure coherence in cognitive modelling.
• Analyse and validate behaviour of autonomous systems in both simulation and field test environments.

Requirements:
• Solid background in reinforcement learning, probabilistic inference, or computational neuroscience.
• Experience with Active Inference, Bayesian inference, or hierarchical generative models.
• Proficiency in Python (PyTorch, TensorFlow, or JAX), with the ability to implement and train complex inference systems.
• Familiarity with decision-making under uncertainty, cognitive architectures, or embodied cognition frameworks.
• Strong theoretical foundation in neuro-inspired AI, behavioural modelling, or theoretical neuroscience.
• Experience integrating sensorimotor control, action selection, or adaptive control in real-time systems.
• Background in robotics, autonomous agents, or AI planning systems is a strong plus.

Note: Experience with interdisciplinary AI combining machine learning, neuroscience, and robotics is highly valued, but not strictly required.

Why you should join us:
• Work in an intellectually stimulating and innovative environment where you can take full ownership of your projects at every stage of development.
• Enjoy flat hierarchies, an open culture, and fast decision-making processes.
• Collaborate with a skilled and dedicated team eager to share their knowledge and expertise.
• Be part of a multinational workplace that values diversity and integrates different backgrounds and perspectives.
• Work in the vibrant heart of Berlin, in the dynamic Kreuzberg district