Astera is a private foundation aiming to steer science and technology toward an abundant future for all. Obelisk, their AGI research lab, is pursuing a neuroscience-informed approach to engineering AGI, unconstrained by funding, profit, or publication needs, and has access to significant computational resources.
Requirements
- Strong background in Reinforcement Learning (RL), autonomous agents, and adaptive learning systems
- Deep knowledge of RL (e.g., model-based RL, meta-RL, hierarchical RL) and experience in designing novel algorithms
- Proficiency in Python and ML frameworks (PyTorch, JAX, TensorFlow)
- Experience with self-play, multi-agent systems, evolutionary algorithms, or life-long learning systems
Responsibilities
- Conduct original research in reinforcement learning and adaptive systems
- Design and implement experiments to test new algorithms and architectures
- Develop agents capable of continual self-improvement without explicit retraining
- Publish results in top AI/ML venues and share insights with the broader research community
- Collaborate with an interdisciplinary team of AI researchers, engineers, and cognitive scientists
Other
- PhD in Computer Science, Machine Learning, AI, Neuroscience, or a closely related field
- Passion for open-ended research, curiosity-driven experimentation, and building systems from first principles
- This role is in-person in Emeryville, CA.
- visa sponsorship may be available for qualified candidates.
