- Define research agendas spanning near-term capabilities, reusable platforms, and field-shaping bets.
- Manage, mentor, and grow researchers with strong scientific taste and technical depth.
- Own the business impact of research by connecting it to product, market, and user needs.
- Frame research theses and design decisive experiments, baselines, evaluations, and failure analyses.
- Engage in model development, post-training, data strategy, and critical implementations.
- Form integrated squads with engineering and domain experts.
- Build compounding assets such as models, datasets, verifiers, benchmarks, and research infrastructure.
- Decide when to deepen, redirect, scale, publish, protect, open-source, or conclude work.
Machine Learning