Agent Evaluation Engineer for an AI Agent Company

DADACONSULTANTS PTE. LTD.; Singapore; On-site; SGD 90,000 - 180,000; Full time

Benefits offered by this job

Frontier AI team

Unlimited AI tools access

Competitive salary

Job summary

DADACONSULTANTS PTE. LTD. is seeking a role focused on evaluating AI agents and models post-training, developing metrics, tasks and experiments to measure performance and value.

You will design evaluation criteria, analyze results, and collaborate with product, engineering and model teams to validate improvements, ensuring robust and unbiased insights for next-gen AI systems.

Qualifications

• Evidence of metric design and validation.
• Understanding of AI model and agent mechanisms.
• Experience in data analysis, experiment design and handling uncertainty.
• Ability to explain findings and their limitations clearly.

Responsibilities

• Design online and offline metrics translating user tasks and output quality into measurable criteria.
• Create evaluation tasks and experiments around agent planning, tool use, context and feedback.
• Support post-training evaluations and compare third-party model quality.
• Develop new evaluation methods for missed issues and test validity and bias.
• Analyze evaluation results and work with product, engineering and model teams to validate improvements.

Skills

Metric design

Experiment design

Data analysis

Agent mechanisms

Task planning

Job description

About our client

Our client is a fast-growing, well-funded AI agent company. The product takes on complex work like deep research, data analysis and software development, and carries it through to a finished result. It is a small team building at the frontier, based in Singapore.

About the role

You will develop evaluations grounded in product needs, user tasks and how AI models and agents actually work. Using metrics, experiments and failure analysis, you'll assess whether capability changes are real, investigate gaps in existing evaluations, and inform system development, post-training and model selection.

What you'll do

• Design online and offline metrics that translate user tasks, output quality and practical value into measurable, testable evaluation criteria
• Design evaluation tasks and experiments around agent planning, tool use, context and feedback, comparing performance before and after system changes
• Support post-training evaluations and comparisons of third-party model quality, defining use cases and the conditions results apply to
• Develop new tasks, metrics or experimental methods for issues existing evaluations miss, and test their validity, bias and reproducibility
• Analyze evaluation results and failures, distinguish score changes from real capability changes, and work with product, engineering and model teams to validate improvements

What you bring

• Practical experience evaluating agent products or model post-training, with concrete evidence of metric design and validation
• Deep understanding of AI model and agent mechanisms — task planning, tool use, context management and feedback
• Strong data analysis, engineering and research skills, including experiment design and handling uncertainty in results
• Ability to investigate open-ended problems, develop well-reasoned new evaluation methods, and examine experimental bias
• Independent judgement on AI output quality and real user value, with the ability to explain findings and their limitations clearly

Nice to have

• Familiarity with user research methods (interviews, observation, usability testing), and the ability to translate findings into evaluation criteria

What we offer

• Build at the frontier of AI agents with a fast-paced team
• Unlimited access to our own AI tools
• Competitive salary, reviewed for the right person