About the role
About Irregular
Irregular is the first frontier security lab with the mission of protecting the world in the time of increasingly capable and sophisticated AI systems. We build next-generation defenses through high-fidelity research platforms that simulate and monitor real-world AI security scenarios.
As AI systems become more powerful, we focus on staying ahead of emerging threats and vulnerabilities, ensuring these transformative technologies can be deployed safely and securely.
About the Role
As a research engineer at Irregular, you'll be at the forefront of building systems to evaluate and secure frontier AI models. You'll work on infrastructure and experiments to assess model capabilities, implement agent frameworks, and develop mitigations for advanced AI systems. Your role will involve creating robust evaluation pipelines, developing security-focused testing frameworks, and building tools that help understand and mitigate risks related to frontier models. You’ll have a chance to understand the research context and your code’s impact and contribute as a meaningful part of a growing team.
Representative projects
- Building a tool to continuously evaluate models and mitigate their risks. From designing the APIs for frontier labs, to building analysis and visualization tools that summarize 10,000+ transcripts into specific conclusions.
- Designing and building challenges that measure a model’s ability to evade discovery, allowing us to see if models can operate on remote systems while avoiding detection by common defensive security tools.
- Developing controlled environment frameworks for more secure use of frontier models.
- Designing and building agents that improve a model’s ability to complete complex tasks. Includes many potential avenues, such as incorporating SOTA prompting practices, creating tools for task delegation, and more.
- Publishing your research and/or delivering research to our customers.
You may be a good fit if you
- Have strong production programming skills and experience.
- Have strong problem-solving and analytical skills.
- Work well in a multidisciplinary team and can adapt to rapidly evolving challenges.
- Are interested in AI and cybersecurity (experience in machine learning or cybersecurity is a plus but not necessary).
- Care about the societal impacts of your work.