AI Research Engineer - AI Safety
Core
Develop cutting-edge techniques for scalable evaluation of AI products, design data collection and experimentation strategies to extract causal insights, and enhance responsible decision-making via uncertainty quantification and safety mechanisms.
Role type
AI Research Engineer (AI Safety & Assurance)
Builds
AI assurance frameworks, evaluation strategies, and safety mechanisms for high-stakes AI systems
Domain
Defense / AI Safety / Machine Learning
Deliverable
production ML models
Required skills
uncertainty quantification, distribution shift detection, adversarial robustness evaluation, deep learning model evaluation, causal inference, software engineering (Python/Rust/C++)
Preferred skills
formal methods, interpretability techniques, Bayesian approaches, adversarial machine learning, red-teaming, conformal prediction, calibration methods
Technologies
Python, Rust, C++, deep learning frameworks
Responsibilities
define operational domains and evaluate reliability of AI capabilities, characterize distribution shifts and failure modes, develop and extend state-of-the-art uncertainty quantification and calibration, design rigorous evaluation frameworks, assess robustness under real-world and adversarial conditions
Seniority
Mid-Senior, hands-on IC