Join a small, technical team in the AI evaluation sector to design and build rigorous benchmarks for frontier AI agents. Your work will help ensure evaluations reflect realistic workflows and provide credible insights into real-world agent performance.
Design, implement, and maintain internal benchmarks for evaluating AI agents on domain-specific tasks.
Work with subject-matter experts to translate real workflows into evaluation tasks and criteria.
Build reliable infrastructure to run models and agents against benchmark tasks at scale.
Develop metrics and analyses to assess benchmark difficulty, reliability, and failure modes.
Validate how benchmark results relate to real-world evaluations and user needs.
Write clear technical documentation and benchmark reports.
Two to four years of experience in software engineering, machine learning engineering, or research, including at least two years building AI benchmarks, evaluation infrastructure, or agent environments.
A degree in a technical field and strong proficiency in Python, Docker, and Linux.
Experience designing and operating benchmark or evaluation environments for AI agents or large language models.
Background building infrastructure to run AI models or agents reliably, and developing metrics, statistical analyses, or validation studies.
Experience collaborating with subject-matter experts to model workflows across technical or business domains and create benchmark tasks.
Strong technical writing, attention to detail, and ability to reason from first principles in an early-stage, unstructured environment.
Published research or technical writing on AI benchmarking, model evaluation, or failure modes is valued. Reinforcement learning experience is also useful.
Base salary: $100,000 to $170,000 USD annually. Visa sponsorship is available.
On-site in Singapore, Singapore.