- 경험
- 어느
- 샐러리
- —
- 채용 공고
- 1
- 게시됨
- 2시간 전
- 작업 모드
- 사무실에서
- 교육
- PhD or equivalent research experience
- 재개하다
- 신청 시 필수 사항
당신이 일하게 될 곳
직무 설명
About the Role
We are collaborating with a distinguished AI research laboratory based in Montreal focused on one of the most critical challenges in artificial intelligence: designing AI systems that are inherently safe. The position is for an ML Research Scientist dedicated to foundational research on safe AI, emphasizing principled, rigorous methods over reactive fixes post-training.
Key Responsibilities
- Conduct foundational research to establish probabilistic safety guarantees for large AI models.
- Create novel approaches for reasoning, causality, robustness, and generalisation in AI.
- Enhance techniques for uncertainty quantification and reliability assessment in language models.
- Contribute to distributed training and rigorous evaluation of large-scale AI models under safety constraints.
- Design and implement strict experimental procedures to verify safety-oriented AI methods.
- Publish influential research papers that drive progress in AI safety.
- Collaborate closely with a small team of top-tier researchers combining theoretical and large-scale practical expertise.
Candidate Profile
- PhD in Computer Science, Physics, Mathematics, or related discipline, or equivalent industrial research experience.
- Track record of publications in prestigious conferences such as NeurIPS, ICML, ICLR, COLT, or ACL.
- Extensive experience with language models and managing large-scale model training workflows.
- Strong theoretical background in probabilistic modeling, Bayesian inference, statistical learning theory, causality, robustness, or uncertainty estimation.
- Robust mathematical skills along with strong production-level programming abilities (proficiency in Python and preferably PyTorch or JAX).
- Experience in designing and conducting large-scale experiments featuring stringent evaluation metrics.
- Pragmatic and realistic perspective on AI safety research, avoiding ideological bias.
- Team-oriented with low ego and a collaborative approach.
Significance
This independent lab operates free from commercial pressures, allowing sustained focus on rigorous and long-term AI safety research. The successful candidate will join a global group of elite researchers dedicated to developing technically sound solutions that can influence the future of safe artificial intelligence.
Location
The position is onsite in Montreal, Quebec, Canada. Candidates willing to relocate are welcomed.