Applied Researcher - Sovereign AI
Core
Build post-training and agent systems for public-sector and regulated-industry settings, operating within strict data, deployment, and access boundaries.
Role type
Applied Researcher (Sovereign AI)
Builds
Post-training pipelines, tool-using domain agents, and high-assurance evaluation frameworks for regulated environments.
Domain
Frontier AI, Sovereign AI, Public Sector, Regulated Industries
Deliverable
production ML models
Required skills
RL/post-training, private ML systems, high-assurance evaluation, agent safety, regulated-domain ML, Python, experimental judgment
Preferred skills
Publications, open-source systems, benchmarks, shipped products
Technologies
Python
Responsibilities
Build private improvement loops that turn sensitive traces into training signals without crossing data boundaries; Train tool-using domain agents with explicit permissions and traceable actions; Develop high-assurance evaluations for reliability, calibration, and compliance; Run post-training experiments against real baselines to attribute gains to model or data.
Seniority
Senior, hands-on IC
