Responsibilities
- Pioneer novel AI architectures, theoretical models, and safety mechanisms in pursuit of a provable framework for safe artificial general intelligence.
- Collaborate directly with Yoshua Bengio and top researchers to define formal safety requirements for AI systems, establish probabilistic safety guarantees under realistic data and computational limits, and support empirical validation as systems evolve.
- Evaluate and expand upon existing theoretical research conducted by the team, contributing to peer-reviewed publications and scientific discourse.
- Maintain deep familiarity with current AI safety and machine learning literature, with a focus on understanding risks such as goal misalignment, reward hacking, and instrumental incentives.
- Help shape and refine the organization’s research direction by identifying high-impact safety questions and guiding experimental priorities without compromising system capability.
- Support internal teams by sharing expertise in AI safety to improve the design and implementation of machine learning algorithms and experiments.
Compensation
Additional compensation totaling 8% of salary allocated for retirement savings or bonuses, independent of performance metrics.
Work Arrangement
On-site — Little Italy, Mile-Ex district
Team
Collaborative and inclusive work environment in a vibrant office space.
Other
- Opportunity to travel regularly with Yoshua Bengio and team to academic and research conferences.
- Comprehensive health benefits package, including mental health support and a wellness management account.
- 20 days of vacation per year from the start date.
- Employer contributes 4% to retirement savings with no required employee match.
- Additional compensation equal to 8% of salary for use in retirement savings or bonuses, regardless of group or individual performance.
- Collaborative and inclusive work environment in a vibrant office space.
