Research Scientist, Foundation Model (World Model & Site Memory)

Laelaps AI is seeking a Research Scientist, Foundation Model (World Model & Site Memory) to lead research on their foundation model of the physical world. The company focuses on developing intelligent software for physical security, enabling robots and sensors to handle critical tasks autonomously. This role is pivotal in advancing the company's mission to create a safer, more secure world through innovative AI solutions. The main day-to-day responsibilities involve developing a probabilistic memory model for each site, determining what constitutes an anomaly based on learned patterns, and creating a world model that predicts future states using field data. The scientist will also work on transferring knowledge from one site's memory to a comprehensive model that applies across multiple sites and domains. Additionally, the role includes designing experiments, choosing research questions, and ensuring that findings are both rigorous and practical for product integration. To qualify for this role, candidates must have a PhD in machine learning, computer vision, robotics, or a related field, with first-author publications in top venues. Experience in world models, self-supervised learning, or probabilistic spatio-temporal modeling is essential. Additionally, candidates should have experience with anomaly or novelty detection in scenarios where positive instances are rare. Strong proficiency in PyTorch and the ability to run experiments at scale are also required. Nice-to-have skills include experience with scene graphs, multi-object tracking, or trajectory forecasting. Knowledge of Bayesian or probabilistic deep learning, robotics, or embodied AI, as well as simulation for data generation, would be beneficial. The role is based in Zurich, offering a collaborative environment with a strong founding team that includes PhD-level co-founders in AI, Robotics, and Physics. The company emphasizes a culture of innovation and fun, with a competitive equity/salary package and opportunities for significant career growth. What the role asks for: - PhD in machine learning, computer vision, robotics or a related field, with first-author publications at top venues. - Research experience in world models, video or latent prediction, self-supervised representation learning, or probabilistic modelling of spatio-temporal data. - Experience with anomaly or novelty detection where positives are rare. - Strong PyTorch and the engineering to run your own experiments at scale. - Nice to have: Scene graphs, multi-object tracking or trajectory forecasting. - Nice to have: Bayesian or probabilistic deep learning. - Nice to have: Robotics or embodied AI. - Nice to have: Simulation for data generation.