Irregular

You'll join a frontier security lab building next-generation AI defenses, owning product features end-to-end, and leading frontend architecture. Your work will turn deep technical capabilities into polished, highly effective user experiences for sophisticated AI organizations.

You'll architect and scale Irregular's security foundations, working with the CISO to execute the security roadmap. You'll design security from the ground up across cloud infrastructure, identity, and application security. This is a high-impact role at a **frontier security lab** protecting against **advanced AI threats**.

You'll build **systems to evaluate and secure frontier AI models** by implementing agent frameworks and developing mitigations. Your work creates **robust evaluation pipelines** and **security-focused testing frameworks** that help understand and mitigate risks. You'll have direct impact on a growing team.

You'll research **frontier model capabilities** and **agent behavior** to shape AI security evaluation and mitigation. Your work will advance knowledge of dangerous capabilities and contribute to cutting-edge mitigations.

You'll help shape **AI security** for cutting-edge products at a **mission-driven company**. Your work will directly protect critical systems. Bring your unique expertise and we'll build the role together.

You'll conduct **AI security research** for a mission-driven company, protecting frontier models from cyber threats. You'll create original evaluation challenges and lead mitigation research, shaping the future of AI security at **the intersection of cybersecurity and AI capabilities**.

You'll tackle the biggest questions in AI development and deployment, shaping the field of **AI security** through research and policy proposals. Your work will directly impact governments, frontier labs, and deployers. You'll produce threat models, taxonomies, and **academic publications** that define safe AI practices.

You'll design and scale production systems for a frontier AI evaluation platform. You'll build distributed pipelines, backend services, and monitoring tools that support **research experiments at scale**. Your work directly measures **AI model security** through automated evaluations.

You'll own and evolve cloud infrastructure and CI/CD pipelines for a fast-growing startup supporting frontier AI labs. You will design scalable systems and **ephemeral, on-demand environments** for leading AI safety research. This role offers high ownership and **real influence over architectural decisions**.