World Model / Action Policy Researcher
World Model / Action Policy Researcher
World Model / Action Policy Researcher
Medal
Medal is seeking a Machine Learning Researcher specializing in world models and action policies to enhance their gaming platform. The role involves designing, training, and evaluating intelligent agents that operate in complex 3D environments, contributing to the next generation of gaming experiences. The position requires a strong background in deep learning and reinforcement learning, with a focus on practical applications in gaming.
Qualification
- 5+ years of experience in deep learning research or reinforcement learning, focusing on embodied agents or simulation environments.
- Strong foundation in representation learning and generative modeling, particularly with diffusion models, VAEs, and transformers applied to video.
- Experience with world models and predictive control for training models that simulate dynamics and plan actions.
- Proficiency in reinforcement learning (RL, model-based RL, or imitation learning) and policy network design.
- Programming fluency in Python and deep learning frameworks such as PyTorch.
Responsibility
- Design, train, and evaluate world models and action policies for gaming applications.
- Collaborate with product and engineering teams to implement research ideas into production.
- Experiment with architectures and iterate rapidly to improve model performance.
- Manage large-scale training and evaluation pipelines for complex datasets or simulations.
- Contribute to the development of intelligent agents that understand and act in 3D environments.
