Research Engineer, Benchmarks

About the Role

This is a hands-on research engineering role focused on designing and owning high-quality benchmarks that evaluate frontier AI agents on realistic, domain-specific workflows. You will sit within a small, highly technical team and play a critical part in ensuring evaluations are rigorous, credible, and trusted by leading AI labs and customers.

What You'll Do

• Design, implement, and own the quality of internal benchmarks for evaluating frontier agents on domain-specific tasks.

• Partner with subject-matter experts to define realistic workflows and translate them into benchmark tasks and evaluation criteria.

• Build and operate reliable infrastructure to run models and agents against benchmark tasks at scale.

• Develop metrics and statistical analyses that measure benchmark difficulty, reliability, and failure modes.

• Validate that benchmark performance correlates with real-world evaluations, customer needs, and frontier lab expectations.

• Write clear technical documentation and benchmark reports for research and engineering audiences.

What We're Looking For

• 2 to 4 years of experience in software engineering, ML engineering, or research roles, with a focused track record in AI benchmarks or evaluation infrastructure.

• Strong proficiency in Python, Docker, and Linux environments.

• Demonstrated experience designing, implementing, and running benchmarks or evaluation environments for AI agents or large language models.

• Experience building infrastructure to reliably run AI models or agents against benchmark or evaluation tasks.

• Ability to analyze and model workflows across diverse technical or business domains to support task design.

• Sharp attention to detail with a habit of spotting subtle inconsistencies and edge cases.

• Comfort reasoning from first principles about task design, scoring, and failure modes.

• Strong written communication skills; experience producing technical documentation or benchmark reports.

• Ability to thrive in unstructured problem spaces at an early-stage startup.

• Bonus: experience with reinforcement learning pipelines, data generation, or RL agent evaluation; published work on AI benchmarking or model evaluation.

Compensation & Benefits

Salary range: USD 150,000 to 250,000 annually. Visa sponsorship is available.

Location

On-site in Singapore.