You will design, build, and maintain cybersecurity benchmarks for frontier AI models within the evaluation infrastructure. Additionally, you will harden evaluation environments to prevent AI escape and use threat modelling to inform the overall benchmarking strategy.
The Talent Acquisition Coordinator will manage the administrative and data backbone of the hiring process, including ATS maintenance, reporting, and candidate communications. They will also support recruiters with scheduling, work test administration, and eventually transition into sourcing and application review.
You will lead the strategy, prioritization, and execution of the company website, ensuring alignment between research goals and user experience. This involves managing the end-to-end project lifecycle, from setting specs and coordinating with design and development teams to performing quality assurance.
The Special Projects Associate will provide specialized operations support for research teams, including managing complex logistics and coordinating with external experts. Responsibilities also include supporting data projects, organizing symposiums, and managing external communication channels.
The specialist will own security protocols, manage access across core platforms, and identify and remediate security risks. They will also partner with engineering and research teams to implement practical security standards without creating unnecessary friction.
The researcher will analyze and critique AI benchmarks every two weeks to evaluate their methodologies and implications for AI capabilities. They will publish public-facing reports and maintain them as new models and benchmark versions are released.
Lead the creation and curation of a benchmark suite to evaluate frontier AI models on real-world office tasks. Design grading rubrics, conduct quantitative and qualitative assessments, and communicate findings through public reports and data visualizations.
Implement and maintain AI benchmarking infrastructure using the Inspect library to evaluate frontier AI models. Collaborate with researchers and analysts to develop new benchmarks and integrate evaluation data into research products.
Lead the UI/UX and data visualization strategy to translate complex research into intuitive digital products and dashboards. Manage the end-to-end design process for the website and mentor other designers and contributors.
