Audits frontier AI systems for deceptive alignment risks

Research Scientist (Control)

Apollo Research is building AI safety products that translate control research into practical tools for reducing risks from AI. The Research Scientist (Control) will design and implement control protocols, conduct empirical experiments, develop monitoring systems and evaluation frameworks, and fine-tune models for detecting coding-agent failures in production environments.

Full-stack Software Engineer (Product)

Apollo Research studies and mitigates risks from advanced AI systems, including deceptive alignment, scheming, and loss of control. The Full-stack Software Engineer will build and scale the Watcher coding agent security product across backend, frontend, monitoring, integrations, infrastructure, and enterprise deployments.

Research Scientist/Engineer (Evaluations)

Apollo Research is focused on assessing the risks posed by scheming AIs, collaborating with leading labs like OpenAI and Google DeepMind. The role involves running evaluations on AI systems, analyzing behavioral patterns, and improving evaluation pipelines.

Research Scientist/Engineer (Science of Scheming)

Apollo Research is focused on the risks associated with AI systems and is looking for Research Scientists and Research Engineers to develop a hard science around the 'Science of Scheming'. The role involves collaborating with AI developers, studying reinforcement learning dynamics, and developing novel evaluation techniques for AI cognition.

Governance Researcher

Apollo Research is a mission-driven startup committed to reducing the threats from scheming frontier AI systems. They are seeking a Governance Researcher to conduct independent, high-quality research on the implications of AI scheming and loss of control threats, contributing to the organization's research agenda and collaborating with other researchers.

Governance Researcher (Expression of Interest)

Backend Software Engineer

Apollo Research is focused on addressing the risks associated with AI capabilities, particularly in the realm of AGI safety. They are seeking a Backend Software Engineer to develop tools for evaluating and monitoring language models, collaborating closely with researchers to enhance their capabilities.

Backend Software Engineer (Research team)

Full-stack Software Engineer

Apollo Research is focused on addressing risks associated with AI capabilities, particularly in AGI safety research. They are seeking a Full-stack Software Engineer to develop tools for analyzing model evaluation results and build libraries for monitoring and controlling LLM traffic.

Full-stack Software Engineer (Research team)

Finance Associate

Apollo Research is an AI safety organization focused on detecting and mitigating scheming. They are seeking a Finance Associate to assist in the day-to-day financial operations and help build the finance team, ensuring accurate financial reporting and management.