Skip to main content

Mathematical Scientist for AI Safety Research

Summary by AIEngineer.careers

Mathematical Scientist for AI Safety Research is a AI Safety Engineer role (full-time). with LawZero. in BERLIN, DE. Imported listing (source: feeny.ai). Apply on the employer's site (feeny.ai).

Imported listing (source: feeny.ai) · Apply on feeny.ai

How our listings work

BERLIN · DE
—
Full-time
SENIOR
Apply Now

More AI jobs in Germany

The sections below reproduce the third-party job description for reference. AIEngineer.careers does not write or control this text.

Imported job description

Sourced from feeny.ai

LawZero, a non-profit founded by Yoshua Bengio, is seeking a theoretical researcher to help design safe-by-design AI systems with probabilistic safety guarantees. The role focuses on formalizing safety requirements, proving guarantees under finite data and compute constraints, and advancing the Scientist AI research program. You will work closely with Yoshua Bengio and a world-class team, contributing to theory, papers, and empirical validation.

Responsibilities

  • Explore new AI system architectures, theoretical frameworks, and safety mechanisms toward a blueprint for safe AGI
  • Formalize requirements and prove probabilistic safety guarantees under finite data and compute constraints
  • Analyze and contribute to existing theoretical work, including writing and disseminating scientific papers
  • Stay current with AI safety and ML research, including misalignment, reward hacking, and instrumental goals
  • Help researchers and engineers design and implement new ML algorithms and experiments

Requirements

  • PhD in machine learning, computer science, mathematics, statistics, or a related area
  • Track record of research, especially formalizing and analyzing subtle natural-language intuition
  • Strong interest and ideally experience in technical AI safety and abstract systems thinking
  • Deep understanding of probabilistic machine learning mathematics for neural networks and language models
  • Clear written and verbal communication to technical and non-technical audiences
  • Ability to work collaboratively on collective research goals

Nice to Have

  • Willingness to regularly travel with Yoshua and team members to conferences

Tech Stack

DPOLarge Language ModelsProbabilistic Machine LearningPyTorchReinforcement LearningRLHFRLVR
LawZero

LawZero

Montréal · CA · 46+ employees

LawZero is a Montreal-based nonprofit startup developing technical solutions for highly capable, safe-by-design AI systems. Its research-led mission is to help ensure advanced AI is developed and deployed responsibly, equitably, and as a global public good.

Weekly job alert

Get a weekly email when new AI engineering jobs match your filters. Free — confirm your address to start.

Related Jobs