Building
Continously evolving systems

The next wave of intelligent systems will autonomously adapt to real-world feedback. We're building adaptive, self-improving agents and their interaction environments, with a mission to design systems for a future with safe superintelligence.

Research

Hard Examples Are All You Need: Maximizing GRPO Post-Training Under Annotation Budgets

Our latest research demonstrates how focusing on challenging examples can significantly improve model performance during post-training, even with limited annotation resources. This work presents novel strategies for efficient data selection and training optimization.

Join Us

We're well-funded ($30M raised) and actively hiring. We prefer in-person in San Francisco, but we are open to high-agency remote team members. Email us at contact@eternis.ai

Footer Image