Applied Scientist

Job Summary

To design and develop methods for enhancing the reliability and safety of multi-agent systems, the full-time remote Applied Scientist will analyze AI agent behavior, fine-tune detection mechanisms, and optimize systems through research insights and collaboration with engineering teams.

Key responsibilities

• Design and develop methods to measure and improve the reliability, safety, and robustness of multi-agent systems
• Analyze the behavior of AI agents under stress using Vijil Diamond and enhance detection capabilities for novel attacks
• Collaborate with research and product engineering teams to translate state-of-the-art research into customer features and contribute to patents and publications

Required qualifications

• Ph.D. in AI/ML, Math, Statistics, or Computer Science
• 4+ years of industry experience as an Applied Scientist or in a similar role
• 2+ years of research experience with large language models (LLMs) or systems for LLM operations
• 1+ year of research experience focusing on agent failure modes
• Demonstrable proficiency in programming with Python, Java, or C++