Research, Post-Training Data

📋 Description

• • Research, Post-Training Data role at Cognition, an applied AI lab building end-to-end software agents like Devin and Windsurf, where you will bridge raw model intelligence with human-useful, safe, and collaborative AI systems.
• • Design and execute data collection and synthesis strategies using human feedback, preference data, and synthetic examples; develop scalable pipelines for human and model-assisted labeling; research human preference modeling to improve reasoning, truthfulness, and helpfulness.
• • Join a small, talent-dense team of world-class competitive programmers, former founders, and researchers from Scale AI, Palantir, Cursor, Google DeepMind, and similar frontier AI organizations, working on AI that acts as a genuine teammate to engineers.
• • Publish and present research that advances the field, share code and datasets, and contribute to scalable oversight and human-AI collaboration paradigms while working with frontier-scale compute and tooling from day one.