Conduct comprehensive failure analysis on AI agent performance across finance-sector tasks to identify patterns, root causes, and systemic issues in the evaluation framework.
Requirements
- Strong foundation in statistical analysis, hypothesis testing, and pattern recognition
- Proficiency in Python (pandas, scipy, matplotlib/seaborn) or R for data analysis
- Experience with exploratory data analysis and creating actionable insights from complex datasets
- Understanding of LLM evaluation methods and quality metrics
- Comfortable working with Excel, data visualization tools (Tableau/Looker), and SQL
- Experience with AI/ML model evaluation or quality assurance
- Familiarity with benchmark datasets and evaluation frameworks
Responsibilities
- Statistical Failure Analysis: Identify patterns in AI agent failures across task components (prompts, rubrics, templates, file types, tags)
- Root Cause Analysis: Determine whether failures stem from task design, rubric clarity, file complexity, or agent limitations
- Dimension Analysis: Analyze performance variations across finance sub-domains, file types, and task categories
- Reporting & Visualization: Create dashboards and reports highlighting failure clusters, edge cases, and improvement opportunities
- Quality Framework: Recommend improvements to task design, rubric structure, and evaluation criteria based on statistical findings
- Stakeholder Communication: Present insights to data labeling experts and technical teams
Other
- Background in finance or willingness to learn finance domain concepts
- Experience with multi-dimensional failure analysis
- 2-4 years of relevant experience
